qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [RFC v18 00/15] i386 cleanup PART 2
@ 2021-02-12 12:36 Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass Claudio Fontana
                   ` (16 more replies)
  0 siblings, 17 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

v17 -> v18:

* meson: add target_user_arch

 - add target_user_arch to all targets which build user.
   Otherwise meson complains about missing key for archs without it.
   (Paolo)

* wrap a few gen_helper_ calls around ifndef CONFIG_USER_ONLY.
  This would need a look from someone like Alex or Richard I think,
  as potentially we could remove even more code I think around the
  gen_helper_ calls for CONFIG_USER_ONLY.

  In the current master code, we have empty helpers for user mode,
  but still we generate the preamble code, temporary variables etc,
  just to then call a helper_() function that does nothing.

  In particular I am referring to patches:

  i386: split tcg btp_helper into softmmu and user parts
        DEF_HELPER_FLAGS_3(set_dr, TCG_CALL_NO_WG, void, env, int, tl)
        DEF_HELPER_FLAGS_4(bpt_io, TCG_CALL_NO_WG, void, env, i32, i32, tl)
        gen_bpt_io
        gen_helper_set_dr(cpu_env, s->tmp2_i32, s->T0);

  i386: split smm helper (softmmu)
        DEF_HELPER_1(rsm, void, env)
	gen_helper_rsm(cpu_env);

  (Alex, Richard?)

* removed suffixes from user/ and softmmu/ modules
  (Alex, Philippe).
  Where possible, removed user stubs entirely.
  Renamed the leftover svm_helper stubs to user/svm_stubs.c

* cleaned up lefover unnecessary header files and squashed them.
 

v16 -> v17: changed to RFC

* tcg_ops are already in master, removed from the series

* i386: split cpu accelerators from cpu.c, using AccelCPUClass:
  removed spurious ; and added spacing before/after functions (Richard)

* added new patches as RFC for the next steps, introducing target-specific
  user-mode specific meson variables, and applied to i386/tcg as an
  example, in order to gather feedback.

v15 -> v16:

* cpu: Move synchronize_from_tb() to tcg_ops:
  - adjusted comments (Alex)

* cpu: tcg_ops: move to tcg-cpu-ops.h, keep a pointer in CPUClass:
  - remove forward decl. of AccelCPUClass, should be in a later patch. (Alex)
  - simplified comment about tcg_ops in struct CPUClass (Alex)
  - remove obsolete comment about ARM blocking TCGCPUOps from being const.
    (Alex)

* accel: replace struct CpusAccel with AccelOpsClass:
  - reworded commit message to be clearer about the objective (Alex)

* accel: introduce AccelCPUClass extending CPUClass
  - reworded commit message to be clearer about the objective (Alex)

* hw/core/cpu: call qemu_init_vcpu in cpu_common_realizefn:
  - dropped this patch (Alex, Philippe)

  will try again later, also in the context of:
  https://www.mail-archive.com/qemu-devel@nongnu.org/msg686480.html

* accel: introduce new accessor functions
  - squashed comments in previous patch introducing accel-cpu.h. (Philippe)

* accel-cpu: make cpu_realizefn return a bool
  - split in two patches, separating the change to the phys_bits check
    (Philippe)

v14 -> v15:

* change the TcgCpuOperations so that all fields of the struct are
  defined unconditionally, as per original patch by Eduardo,
  before moving them to a separate struct referenced by a pointer
  (Richard, Eduardo).

* changed (c) year to 2021

* added a patch to make accel_cpu->cpu_realizefn return bool, and
  adapt host_cpu, kvm_cpu, hvf_cpu and tcg_cpu in i386 accordingly.
  Ultimately, consistently moving to a pattern of realize functions
  returning bool will require a rework of all devices.

v13 -> v14: rebased on latest master.
v12 -> v13: rebased on latest master.

v11 -> v12: reordered patches and improved tcg_ops

* reordered all TcgCpuOperations stuff so it is at the beginning

* added patches for ARM-specific tcg ops
  debug_check_watchpoint and adjust_watchpoint_address

* added a patch that puts a forward declared pointer in the struct,
  so as to reduce the change of misuse between common_ss and specific_ss code,
  and tidy up as a consequence all targets, by defining dedicated structs.

v10 -> v11: split off PART 2,

no further changes to PART 2 other than the split.

v9 -> v10: minor tweaks and fixes

* in "i386: split cpu accelerators from cpu.c",

use kvm/kvm-cpu.c, hvf/hvf-cpu.c, tcg/tcg-cpu.c.
Easier to understand compared to editing multiple cpu.c files,
and matches the header files if needed (kvm-cpu.h).

* in "accel: replace struct CpusAccel with AccelOpsClass",

make it a bit more consistent, by naming the files defining
the AccelOpsClass types "...-accel-ops.c" instead of the old
naming "...-cpus.c".

* in "cpu: move cc->transaction_failed to tcg_ops",

protect with CONFIG_TCG the use of tcg_ops for hw/misc/jazz.c,

 #include "exec/memattrs.h" (Philippe, Eduardo)

* in "cpu: Move synchronize_from_tb() to tcg_ops",

 #include "hw/core/cpu.h" (Philippe, Eduardo)

do not remove the comment about struct TcgCpuOperations (Philippe)

* in "accel/tcg: split TCG-only code from cpu_exec_realizefn",

invert tcg_target_initialized set order (Alex)

* in "i386: move TCG cpu class initialization out of helper.c",

extract helper-tcg.h, tcg-cpu.c, and tcg-cpu.h directly into
tcg/, avoiding the extra move later to tcg/ (Alex)



v8 -> v9: move additional methods to CPUClass->tcg_ops

do_unaligned_access, transaction_failed and do_interrupt.

do_interrupt is a bit tricky, as the same code is reused
(albeit not usually directly) for KVM under certain odd conditions.

Change arm, as the only user of do_interrupt callback for KVM,
to instead call the target function directly arm_do_interrupt.

v7 -> v8: add missing CONFIG_TCGs, fix bugs

* add the prerequisite patches for "3 tcg" at the beginning of the
  series for convenience (already reviewed, queued by RH).

* add CONFIG_TCG to TCGCpuOperations and tcg_ops variable use

* reduce the scope of the realizefn refactoring, do not
  introduce a separate cpu_accel_realize, and instead use the
  existing cpu_exec_realizefn, there is not enough benefit
  to introduce a new function.

* fix bugs in user mode due to attempt to move the tcg_region_init()
  early, so it could be done just once in tcg_init() for both
  softmmu and user mode. Unfortunately it needs to remain deferred
  for user mode, as it needs to be done after prologue init and
  after the GUEST_BASE has been set.

v6 -> v7: integrate TCGCpuOperations, refactored cpu_exec_realizefn

* integrate TCGCpuOperations (Eduardo)

Taken some refactoring from Eduardo for Tcg-only operations on
CPUClass.

* refactored cpu_exec_realizefn

The other main change is a refactoring of cpu_exec_realizefn,
directly linked to the effort of making many cpu_exec operations
TCG-only (Eduardo series above):

cpu_exec_realizefn is actually a TCG-only thing, with the
exception of a couple things that can be done in base cpu code.

This changes all targets realizefn, so I guess I have to Cc:
the Multiverse? (Universe was already CCed for all accelerators).


v5 -> v6: remove MODULE_INIT_ACCEL_CPU


instead, use a call to accel_init_interfaces().

* The class lookups are now general and performed in accel/

  new AccelCPUClass for new archs are supported as new
  ones appear in the class hierarchy, no need for stubs.

* Split the code a bit better


v4 -> v5: centralized and simplified initializations

I put in Cc: Emilio G. Cota, specifically because in patch 8
I (re)moved for user-mode the call to tcg_regions_init().

The call happens now inside the tcg AccelClass machine_init,
(so earlier). This seems to work fine, but thought to get the
author opinion on this.

Rebased on "tcg-cpus: split into 3 tcg variants" series
(queued by Richard), to avoid some code churn:


https://lists.gnu.org/archive/html/qemu-devel/2020-10/msg04356.html


* Extended AccelClass to user-mode.

user-mode now does not call tcg_exec_init directly,
instead it uses the tcg accel class, and its init_machine method.

Since user-mode does not define or use a machine state,
the machine is just passed as NULL.

The immediate advantage is that now we can call current_accel()
from both user mode and softmmu, so we can work out the correct
class to use for accelerator initializations.

* QOMification of CpusAccelOps

simple QOMification of CpusAccelOps abstract class.

* Centralized all accel_cpu_init, so only one per cpu-arch,
  plus one for all accels will remain.

  So we can expect accel_cpu_init() to be limited to:
  
  softmmu/cpus.c - initializes the chosen softmmu accel ops for the cpus module.
  target/ARCH/cpu.c - initializes the chosen arch-specific cpu accelerator.
  
These changes are meant to address concerns/issues (Paolo):

1) the use of if (tcg_enabled()) and similar in the module_init call path

2) the excessive number of accel_cpu_init() to hunt down in the codebase.


* Fixed wrong use of host_cpu_class_init (Eduardo)


v3 -> v4: QOMification of X86CPUAccelClass


In this version I basically QOMified X86CPUAccel, taking the
suggestions from Eduardo as the starting point,
but stopping just short of making it an actual QOM interface,
using a plain abstract class, and then subclasses for the
actual objects.

Initialization is still using the existing qemu initialization
framework (module_call_init), which is I still think is better
than the alternatives proposed, in the current state.

Possibly some improvements could be developed in the future here.
In this case, effort should be put in keeping things extendible,
in order not to be blocked once accelerators also become modules.

Motivation and higher level steps:

https://lists.gnu.org/archive/html/qemu-devel/2020-05/msg04628.html

Looking forward to your comments on this proposal,

Ciao,

Claudio


Claudio Fontana (15):
  i386: split cpu accelerators from cpu.c, using AccelCPUClass
  cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn
  accel: introduce new accessor functions
  target/i386: fix host_cpu_adjust_phys_bits error handling
  accel-cpu: make cpu_realizefn return a bool
  meson: add target_user_arch
  i386: split off softmmu-only functionality in tcg-cpu
  i386: split smm helper (softmmu)
  i386: split tcg excp_helper into softmmu and user parts
  i386: split tcg btp_helper into softmmu and user parts
  i386: split misc helper into user and softmmu parts
  i386: separate fpu_helper into user and softmmu parts
  i386: slit svm_helper into softmmu and stub-only user
  i386: split seg_helper into user-only and softmmu parts
  i386: split off softmmu part of cpu.c

 meson.build                                |   5 +
 include/hw/core/accel-cpu.h                |   2 +-
 include/qemu/accel.h                       |  13 +
 target/i386/cpu-internal.h                 |  70 ++
 target/i386/cpu.h                          |  23 +-
 target/i386/helper.h                       |  11 +
 target/i386/host-cpu.h                     |  19 +
 target/i386/kvm/kvm-cpu.h                  |  41 ++
 target/i386/tcg/helper-tcg.h               |   8 +
 target/i386/tcg/seg_helper.h               |  66 ++
 target/i386/tcg/tcg-cpu.h                  |  21 +-
 accel/accel-common.c                       |  19 +
 cpu.c                                      |   5 +-
 hw/i386/pc_piix.c                          |   1 +
 target/i386/cpu-softmmu.c                  | 352 ++++++++++
 target/i386/cpu.c                          | 780 ++-------------------
 target/i386/host-cpu.c                     | 204 ++++++
 target/i386/hvf/hvf-cpu.c                  |  68 ++
 target/i386/kvm/kvm-cpu.c                  | 151 ++++
 target/i386/kvm/kvm.c                      |   3 +-
 target/i386/tcg/bpt_helper.c               | 276 --------
 target/i386/tcg/excp_helper.c              | 572 ---------------
 target/i386/tcg/fpu_helper.c               |  65 +-
 target/i386/tcg/misc_helper.c              | 463 ------------
 target/i386/tcg/seg_helper.c               | 235 +------
 target/i386/tcg/softmmu/bpt_helper.c       | 293 ++++++++
 target/i386/tcg/softmmu/excp_helper.c      | 562 +++++++++++++++
 target/i386/tcg/softmmu/fpu_helper.c       |  58 ++
 target/i386/tcg/softmmu/misc_helper.c      | 438 ++++++++++++
 target/i386/tcg/softmmu/seg_helper.c       | 125 ++++
 target/i386/tcg/{ => softmmu}/smm_helper.c |  19 +-
 target/i386/tcg/{ => softmmu}/svm_helper.c |  62 +-
 target/i386/tcg/softmmu/tcg-cpu.c          |  83 +++
 target/i386/tcg/tcg-cpu.c                  |  50 +-
 target/i386/tcg/translate.c                |   6 +
 target/i386/tcg/user/excp_helper.c         |  39 ++
 target/i386/tcg/user/fpu_helper.c          |  42 ++
 target/i386/tcg/user/misc_helper.c         |  72 ++
 target/i386/tcg/user/seg_helper.c          | 109 +++
 target/i386/tcg/user/svm_stubs.c           |  76 ++
 MAINTAINERS                                |   2 +-
 target/alpha/meson.build                   |   3 +
 target/arm/meson.build                     |   2 +
 target/cris/meson.build                    |   3 +
 target/hppa/meson.build                    |   3 +
 target/i386/hvf/meson.build                |   1 +
 target/i386/kvm/meson.build                |   7 +-
 target/i386/meson.build                    |   9 +-
 target/i386/tcg/meson.build                |   5 +-
 target/i386/tcg/softmmu/meson.build        |  10 +
 target/i386/tcg/user/meson.build           |   7 +
 target/m68k/meson.build                    |   3 +
 target/microblaze/meson.build              |   3 +
 target/mips/meson.build                    |   3 +
 target/nios2/meson.build                   |   3 +
 target/openrisc/meson.build                |   3 +
 target/ppc/meson.build                     |   3 +
 target/riscv/meson.build                   |   3 +
 target/s390x/meson.build                   |   3 +
 target/sh4/meson.build                     |   3 +
 target/sparc/meson.build                   |   3 +
 target/tilegx/meson.build                  |   3 +
 target/tricore/meson.build                 |   3 +
 target/xtensa/meson.build                  |   3 +
 64 files changed, 3148 insertions(+), 2450 deletions(-)
 create mode 100644 target/i386/cpu-internal.h
 create mode 100644 target/i386/host-cpu.h
 create mode 100644 target/i386/kvm/kvm-cpu.h
 create mode 100644 target/i386/tcg/seg_helper.h
 create mode 100644 target/i386/cpu-softmmu.c
 create mode 100644 target/i386/host-cpu.c
 create mode 100644 target/i386/hvf/hvf-cpu.c
 create mode 100644 target/i386/kvm/kvm-cpu.c
 create mode 100644 target/i386/tcg/softmmu/bpt_helper.c
 create mode 100644 target/i386/tcg/softmmu/excp_helper.c
 create mode 100644 target/i386/tcg/softmmu/fpu_helper.c
 create mode 100644 target/i386/tcg/softmmu/misc_helper.c
 create mode 100644 target/i386/tcg/softmmu/seg_helper.c
 rename target/i386/tcg/{ => softmmu}/smm_helper.c (98%)
 rename target/i386/tcg/{ => softmmu}/svm_helper.c (96%)
 create mode 100644 target/i386/tcg/softmmu/tcg-cpu.c
 create mode 100644 target/i386/tcg/user/excp_helper.c
 create mode 100644 target/i386/tcg/user/fpu_helper.c
 create mode 100644 target/i386/tcg/user/misc_helper.c
 create mode 100644 target/i386/tcg/user/seg_helper.c
 create mode 100644 target/i386/tcg/user/svm_stubs.c
 create mode 100644 target/i386/tcg/softmmu/meson.build
 create mode 100644 target/i386/tcg/user/meson.build

-- 
2.26.2



^ permalink raw reply	[flat|nested] 33+ messages in thread

* [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:29   ` Alex Bennée
  2021-02-12 12:36 ` [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn Claudio Fontana
                   ` (15 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

i386 is the first user of AccelCPUClass, allowing to split
cpu.c into:

cpu.c            cpuid and common x86 cpu functionality
host-cpu.c       host x86 cpu functions and "host" cpu type
kvm/kvm-cpu.c    KVM x86 AccelCPUClass
hvf/hvf-cpu.c    HVF x86 AccelCPUClass
tcg/tcg-cpu.c    TCG x86 AccelCPUClass

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/cpu.h           |  20 +-
 target/i386/host-cpu.h      |  19 ++
 target/i386/kvm/kvm-cpu.h   |  41 ++++
 target/i386/tcg/tcg-cpu.h   |  15 --
 hw/i386/pc_piix.c           |   1 +
 target/i386/cpu.c           | 390 ++++--------------------------------
 target/i386/host-cpu.c      | 201 +++++++++++++++++++
 target/i386/hvf/hvf-cpu.c   |  68 +++++++
 target/i386/kvm/kvm-cpu.c   | 151 ++++++++++++++
 target/i386/kvm/kvm.c       |   3 +-
 target/i386/tcg/tcg-cpu.c   | 113 ++++++++++-
 MAINTAINERS                 |   2 +-
 target/i386/hvf/meson.build |   1 +
 target/i386/kvm/meson.build |   7 +-
 target/i386/meson.build     |   6 +-
 15 files changed, 651 insertions(+), 387 deletions(-)
 create mode 100644 target/i386/host-cpu.h
 create mode 100644 target/i386/kvm/kvm-cpu.h
 delete mode 100644 target/i386/tcg/tcg-cpu.h
 create mode 100644 target/i386/host-cpu.c
 create mode 100644 target/i386/hvf/hvf-cpu.c
 create mode 100644 target/i386/kvm/kvm-cpu.c

diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8d599bb5b8..a4f9bbef55 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1920,13 +1920,20 @@ int cpu_x86_signal_handler(int host_signum, void *pinfo,
                            void *puc);
 
 /* cpu.c */
+void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1,
+                              uint32_t vendor2, uint32_t vendor3);
+typedef struct PropValue {
+    const char *prop, *value;
+} PropValue;
+void x86_cpu_apply_props(X86CPU *cpu, PropValue *props);
+
+/* cpu.c other functions (cpuid) */
 void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                    uint32_t *eax, uint32_t *ebx,
                    uint32_t *ecx, uint32_t *edx);
 void cpu_clear_apic_feature(CPUX86State *env);
 void host_cpuid(uint32_t function, uint32_t count,
                 uint32_t *eax, uint32_t *ebx, uint32_t *ecx, uint32_t *edx);
-void host_vendor_fms(char *vendor, int *family, int *model, int *stepping);
 
 /* helper.c */
 void x86_cpu_set_a20(X86CPU *cpu, int a20_state);
@@ -2131,17 +2138,6 @@ void cpu_report_tpr_access(CPUX86State *env, TPRAccess access);
 void apic_handle_tpr_access_report(DeviceState *d, target_ulong ip,
                                    TPRAccess access);
 
-
-/* Change the value of a KVM-specific default
- *
- * If value is NULL, no default will be set and the original
- * value from the CPU model table will be kept.
- *
- * It is valid to call this function only for properties that
- * are already present in the kvm_default_props table.
- */
-void x86_cpu_change_kvm_default(const char *prop, const char *value);
-
 /* Special values for X86CPUVersion: */
 
 /* Resolve to latest CPU version */
diff --git a/target/i386/host-cpu.h b/target/i386/host-cpu.h
new file mode 100644
index 0000000000..b47bc0943f
--- /dev/null
+++ b/target/i386/host-cpu.h
@@ -0,0 +1,19 @@
+/*
+ * x86 host CPU type initialization and host CPU functions
+ *
+ * Copyright 2021 SUSE LLC
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#ifndef HOST_CPU_H
+#define HOST_CPU_H
+
+void host_cpu_instance_init(X86CPU *cpu);
+void host_cpu_max_instance_init(X86CPU *cpu);
+void host_cpu_realizefn(CPUState *cs, Error **errp);
+
+void host_cpu_vendor_fms(char *vendor, int *family, int *model, int *stepping);
+
+#endif /* HOST_CPU_H */
diff --git a/target/i386/kvm/kvm-cpu.h b/target/i386/kvm/kvm-cpu.h
new file mode 100644
index 0000000000..e858ca21e5
--- /dev/null
+++ b/target/i386/kvm/kvm-cpu.h
@@ -0,0 +1,41 @@
+/*
+ * i386 KVM CPU type and functions
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef KVM_CPU_H
+#define KVM_CPU_H
+
+#ifdef CONFIG_KVM
+/*
+ * Change the value of a KVM-specific default
+ *
+ * If value is NULL, no default will be set and the original
+ * value from the CPU model table will be kept.
+ *
+ * It is valid to call this function only for properties that
+ * are already present in the kvm_default_props table.
+ */
+void x86_cpu_change_kvm_default(const char *prop, const char *value);
+
+#else /* !CONFIG_KVM */
+
+#define x86_cpu_change_kvm_default(a, b)
+
+#endif /* CONFIG_KVM */
+
+#endif /* KVM_CPU_H */
diff --git a/target/i386/tcg/tcg-cpu.h b/target/i386/tcg/tcg-cpu.h
deleted file mode 100644
index 81f02e562e..0000000000
--- a/target/i386/tcg/tcg-cpu.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/*
- * i386 TCG CPU class initialization
- *
- * Copyright 2020 SUSE LLC
- *
- * This work is licensed under the terms of the GNU GPL, version 2 or later.
- * See the COPYING file in the top-level directory.
- */
-
-#ifndef TCG_CPU_H
-#define TCG_CPU_H
-
-void tcg_cpu_common_class_init(CPUClass *cc);
-
-#endif /* TCG_CPU_H */
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 2904b40163..7772880c04 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -64,6 +64,7 @@
 #include "hw/hyperv/vmbus-bridge.h"
 #include "hw/mem/nvdimm.h"
 #include "hw/i386/acpi-build.h"
+#include "kvm/kvm-cpu.h"
 
 #define MAX_IDE_BUS 2
 
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 9c3d2d60b7..b034571869 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -22,38 +22,25 @@
 #include "qemu/cutils.h"
 #include "qemu/bitops.h"
 #include "qemu/qemu-print.h"
-
 #include "cpu.h"
-#include "tcg/tcg-cpu.h"
 #include "tcg/helper-tcg.h"
 #include "exec/exec-all.h"
 #include "sysemu/kvm.h"
 #include "sysemu/reset.h"
 #include "sysemu/hvf.h"
-#include "sysemu/cpus.h"
+#include "hw/core/accel-cpu.h"
 #include "sysemu/xen.h"
 #include "sysemu/whpx.h"
 #include "kvm/kvm_i386.h"
 #include "sev_i386.h"
-
-#include "qemu/error-report.h"
 #include "qemu/module.h"
-#include "qemu/option.h"
-#include "qemu/config-file.h"
-#include "qapi/error.h"
 #include "qapi/qapi-visit-machine.h"
 #include "qapi/qapi-visit-run-state.h"
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qerror.h"
-#include "qapi/visitor.h"
 #include "qom/qom-qobject.h"
-#include "sysemu/arch_init.h"
 #include "qapi/qapi-commands-machine-target.h"
-
 #include "standard-headers/asm-x86/kvm_para.h"
-
-#include "sysemu/sysemu.h"
-#include "sysemu/tcg.h"
 #include "hw/qdev-properties.h"
 #include "hw/i386/topology.h"
 #ifndef CONFIG_USER_ONLY
@@ -595,8 +582,8 @@ static CPUCacheInfo legacy_l3_cache = {
 #define INTEL_PT_CYCLE_BITMAP    0x1fff         /* Support 0,2^(0~11) */
 #define INTEL_PT_PSB_BITMAP      (0x003f << 16) /* Support 2K,4K,8K,16K,32K,64K */
 
-static void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1,
-                                     uint32_t vendor2, uint32_t vendor3)
+void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1,
+                              uint32_t vendor2, uint32_t vendor3)
 {
     int i;
     for (i = 0; i < 4; i++) {
@@ -1589,25 +1576,6 @@ void host_cpuid(uint32_t function, uint32_t count,
         *edx = vec[3];
 }
 
-void host_vendor_fms(char *vendor, int *family, int *model, int *stepping)
-{
-    uint32_t eax, ebx, ecx, edx;
-
-    host_cpuid(0x0, 0, &eax, &ebx, &ecx, &edx);
-    x86_cpu_vendor_words2str(vendor, ebx, edx, ecx);
-
-    host_cpuid(0x1, 0, &eax, &ebx, &ecx, &edx);
-    if (family) {
-        *family = ((eax >> 8) & 0x0F) + ((eax >> 20) & 0xFF);
-    }
-    if (model) {
-        *model = ((eax >> 4) & 0x0F) | ((eax & 0xF0000) >> 12);
-    }
-    if (stepping) {
-        *stepping = eax & 0x0F;
-    }
-}
-
 /* CPU class name definitions: */
 
 /* Return type name for a given CPU model name
@@ -1632,10 +1600,6 @@ static char *x86_cpu_class_get_model_name(X86CPUClass *cc)
                      strlen(class_name) - strlen(X86_CPU_TYPE_SUFFIX));
 }
 
-typedef struct PropValue {
-    const char *prop, *value;
-} PropValue;
-
 typedef struct X86CPUVersionDefinition {
     X86CPUVersion version;
     const char *alias;
@@ -4132,32 +4096,6 @@ static X86CPUDefinition builtin_x86_defs[] = {
     },
 };
 
-/* KVM-specific features that are automatically added/removed
- * from all CPU models when KVM is enabled.
- */
-static PropValue kvm_default_props[] = {
-    { "kvmclock", "on" },
-    { "kvm-nopiodelay", "on" },
-    { "kvm-asyncpf", "on" },
-    { "kvm-steal-time", "on" },
-    { "kvm-pv-eoi", "on" },
-    { "kvmclock-stable-bit", "on" },
-    { "x2apic", "on" },
-    { "kvm-msi-ext-dest-id", "off" },
-    { "acpi", "off" },
-    { "monitor", "off" },
-    { "svm", "off" },
-    { NULL, NULL },
-};
-
-/* TCG-specific defaults that override all CPU models when using TCG
- */
-static PropValue tcg_default_props[] = {
-    { "vme", "off" },
-    { NULL, NULL },
-};
-
-
 /*
  * We resolve CPU model aliases using -v1 when using "-machine
  * none", but this is just for compatibility while libvirt isn't
@@ -4199,61 +4137,6 @@ static X86CPUVersion x86_cpu_model_resolve_version(const X86CPUModel *model)
     return v;
 }
 
-void x86_cpu_change_kvm_default(const char *prop, const char *value)
-{
-    PropValue *pv;
-    for (pv = kvm_default_props; pv->prop; pv++) {
-        if (!strcmp(pv->prop, prop)) {
-            pv->value = value;
-            break;
-        }
-    }
-
-    /* It is valid to call this function only for properties that
-     * are already present in the kvm_default_props table.
-     */
-    assert(pv->prop);
-}
-
-static bool lmce_supported(void)
-{
-    uint64_t mce_cap = 0;
-
-#ifdef CONFIG_KVM
-    if (kvm_ioctl(kvm_state, KVM_X86_GET_MCE_CAP_SUPPORTED, &mce_cap) < 0) {
-        return false;
-    }
-#endif
-
-    return !!(mce_cap & MCG_LMCE_P);
-}
-
-#define CPUID_MODEL_ID_SZ 48
-
-/**
- * cpu_x86_fill_model_id:
- * Get CPUID model ID string from host CPU.
- *
- * @str should have at least CPUID_MODEL_ID_SZ bytes
- *
- * The function does NOT add a null terminator to the string
- * automatically.
- */
-static int cpu_x86_fill_model_id(char *str)
-{
-    uint32_t eax = 0, ebx = 0, ecx = 0, edx = 0;
-    int i;
-
-    for (i = 0; i < 3; i++) {
-        host_cpuid(0x80000002 + i, 0, &eax, &ebx, &ecx, &edx);
-        memcpy(str + i * 16 +  0, &eax, 4);
-        memcpy(str + i * 16 +  4, &ebx, 4);
-        memcpy(str + i * 16 +  8, &ecx, 4);
-        memcpy(str + i * 16 + 12, &edx, 4);
-    }
-    return 0;
-}
-
 static Property max_x86_cpu_properties[] = {
     DEFINE_PROP_BOOL("migratable", X86CPU, migratable, true),
     DEFINE_PROP_BOOL("host-cache-info", X86CPU, cache_info_passthrough, false),
@@ -4276,62 +4159,25 @@ static void max_x86_cpu_class_init(ObjectClass *oc, void *data)
 static void max_x86_cpu_initfn(Object *obj)
 {
     X86CPU *cpu = X86_CPU(obj);
-    CPUX86State *env = &cpu->env;
-    KVMState *s = kvm_state;
 
     /* We can't fill the features array here because we don't know yet if
      * "migratable" is true or false.
      */
     cpu->max_features = true;
-
-    if (accel_uses_host_cpuid()) {
-        char vendor[CPUID_VENDOR_SZ + 1] = { 0 };
-        char model_id[CPUID_MODEL_ID_SZ + 1] = { 0 };
-        int family, model, stepping;
-
-        host_vendor_fms(vendor, &family, &model, &stepping);
-        cpu_x86_fill_model_id(model_id);
-
-        object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
-        object_property_set_int(OBJECT(cpu), "family", family, &error_abort);
-        object_property_set_int(OBJECT(cpu), "model", model, &error_abort);
-        object_property_set_int(OBJECT(cpu), "stepping", stepping,
-                                &error_abort);
-        object_property_set_str(OBJECT(cpu), "model-id", model_id,
-                                &error_abort);
-
-        if (kvm_enabled()) {
-            env->cpuid_min_level =
-                kvm_arch_get_supported_cpuid(s, 0x0, 0, R_EAX);
-            env->cpuid_min_xlevel =
-                kvm_arch_get_supported_cpuid(s, 0x80000000, 0, R_EAX);
-            env->cpuid_min_xlevel2 =
-                kvm_arch_get_supported_cpuid(s, 0xC0000000, 0, R_EAX);
-        } else {
-            env->cpuid_min_level =
-                hvf_get_supported_cpuid(0x0, 0, R_EAX);
-            env->cpuid_min_xlevel =
-                hvf_get_supported_cpuid(0x80000000, 0, R_EAX);
-            env->cpuid_min_xlevel2 =
-                hvf_get_supported_cpuid(0xC0000000, 0, R_EAX);
-        }
-
-        if (lmce_supported()) {
-            object_property_set_bool(OBJECT(cpu), "lmce", true, &error_abort);
-        }
-        object_property_set_bool(OBJECT(cpu), "host-phys-bits", true, &error_abort);
-    } else {
-        object_property_set_str(OBJECT(cpu), "vendor", CPUID_VENDOR_AMD,
-                                &error_abort);
-        object_property_set_int(OBJECT(cpu), "family", 6, &error_abort);
-        object_property_set_int(OBJECT(cpu), "model", 6, &error_abort);
-        object_property_set_int(OBJECT(cpu), "stepping", 3, &error_abort);
-        object_property_set_str(OBJECT(cpu), "model-id",
-                                "QEMU TCG CPU version " QEMU_HW_VERSION,
-                                &error_abort);
-    }
-
     object_property_set_bool(OBJECT(cpu), "pmu", true, &error_abort);
+
+    /*
+     * these defaults are used for TCG and all other accelerators
+     * besides KVM and HVF, which overwrite these values
+     */
+    object_property_set_str(OBJECT(cpu), "vendor", CPUID_VENDOR_AMD,
+                            &error_abort);
+    object_property_set_int(OBJECT(cpu), "family", 6, &error_abort);
+    object_property_set_int(OBJECT(cpu), "model", 6, &error_abort);
+    object_property_set_int(OBJECT(cpu), "stepping", 3, &error_abort);
+    object_property_set_str(OBJECT(cpu), "model-id",
+                            "QEMU TCG CPU version " QEMU_HW_VERSION,
+                            &error_abort);
 }
 
 static const TypeInfo max_x86_cpu_type_info = {
@@ -4341,31 +4187,6 @@ static const TypeInfo max_x86_cpu_type_info = {
     .class_init = max_x86_cpu_class_init,
 };
 
-#if defined(CONFIG_KVM) || defined(CONFIG_HVF)
-static void host_x86_cpu_class_init(ObjectClass *oc, void *data)
-{
-    X86CPUClass *xcc = X86_CPU_CLASS(oc);
-
-    xcc->host_cpuid_required = true;
-    xcc->ordering = 8;
-
-#if defined(CONFIG_KVM)
-    xcc->model_description =
-        "KVM processor with all supported host features ";
-#elif defined(CONFIG_HVF)
-    xcc->model_description =
-        "HVF processor with all supported host features ";
-#endif
-}
-
-static const TypeInfo host_x86_cpu_type_info = {
-    .name = X86_CPU_TYPE_NAME("host"),
-    .parent = X86_CPU_TYPE_NAME("max"),
-    .class_init = host_x86_cpu_class_init,
-};
-
-#endif
-
 static char *feature_word_description(FeatureWordInfo *f, uint32_t bit)
 {
     assert(f->type == CPUID_FEATURE_WORD || f->type == MSR_FEATURE_WORD);
@@ -5084,7 +4905,7 @@ static uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
     return r;
 }
 
-static void x86_cpu_apply_props(X86CPU *cpu, PropValue *props)
+void x86_cpu_apply_props(X86CPU *cpu, PropValue *props)
 {
     PropValue *pv;
     for (pv = props; pv->prop; pv++) {
@@ -5131,8 +4952,6 @@ static void x86_cpu_load_model(X86CPU *cpu, X86CPUModel *model)
 {
     X86CPUDefinition *def = model->cpudef;
     CPUX86State *env = &cpu->env;
-    const char *vendor;
-    char host_vendor[CPUID_VENDOR_SZ + 1];
     FeatureWord w;
 
     /*NOTE: any property set by this function should be returned by
@@ -5159,20 +4978,6 @@ static void x86_cpu_load_model(X86CPU *cpu, X86CPUModel *model)
     /* legacy-cache defaults to 'off' if CPU model provides cache info */
     cpu->legacy_cache = !def->cache_info;
 
-    /* Special cases not set in the X86CPUDefinition structs: */
-    /* TODO: in-kernel irqchip for hvf */
-    if (kvm_enabled()) {
-        if (!kvm_irqchip_in_kernel()) {
-            x86_cpu_change_kvm_default("x2apic", "off");
-        } else if (kvm_irqchip_is_split() && kvm_enable_x2apic()) {
-            x86_cpu_change_kvm_default("kvm-msi-ext-dest-id", "on");
-        }
-
-        x86_cpu_apply_props(cpu, kvm_default_props);
-    } else if (tcg_enabled()) {
-        x86_cpu_apply_props(cpu, tcg_default_props);
-    }
-
     env->features[FEAT_1_ECX] |= CPUID_EXT_HYPERVISOR;
 
     /* sysenter isn't supported in compatibility mode on AMD,
@@ -5182,15 +4987,12 @@ static void x86_cpu_load_model(X86CPU *cpu, X86CPUModel *model)
      * KVM's sysenter/syscall emulation in compatibility mode and
      * when doing cross vendor migration
      */
-    vendor = def->vendor;
-    if (accel_uses_host_cpuid()) {
-        uint32_t  ebx = 0, ecx = 0, edx = 0;
-        host_cpuid(0, 0, NULL, &ebx, &ecx, &edx);
-        x86_cpu_vendor_words2str(host_vendor, ebx, edx, ecx);
-        vendor = host_vendor;
-    }
 
-    object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
+    /*
+     * vendor property is set here but then overloaded with the
+     * host cpu vendor for KVM and HVF.
+     */
+    object_property_set_str(OBJECT(cpu), "vendor", def->vendor, &error_abort);
 
     x86_cpu_apply_version_props(cpu, model);
 
@@ -6220,53 +6022,12 @@ static void x86_cpu_apic_realize(X86CPU *cpu, Error **errp)
         apic_mmio_map_once = true;
      }
 }
-
-static void x86_cpu_machine_done(Notifier *n, void *unused)
-{
-    X86CPU *cpu = container_of(n, X86CPU, machine_done);
-    MemoryRegion *smram =
-        (MemoryRegion *) object_resolve_path("/machine/smram", NULL);
-
-    if (smram) {
-        cpu->smram = g_new(MemoryRegion, 1);
-        memory_region_init_alias(cpu->smram, OBJECT(cpu), "smram",
-                                 smram, 0, 4 * GiB);
-        memory_region_set_enabled(cpu->smram, true);
-        memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->smram, 1);
-    }
-}
 #else
 static void x86_cpu_apic_realize(X86CPU *cpu, Error **errp)
 {
 }
 #endif
 
-/* Note: Only safe for use on x86(-64) hosts */
-static uint32_t x86_host_phys_bits(void)
-{
-    uint32_t eax;
-    uint32_t host_phys_bits;
-
-    host_cpuid(0x80000000, 0, &eax, NULL, NULL, NULL);
-    if (eax >= 0x80000008) {
-        host_cpuid(0x80000008, 0, &eax, NULL, NULL, NULL);
-        /* Note: According to AMD doc 25481 rev 2.34 they have a field
-         * at 23:16 that can specify a maximum physical address bits for
-         * the guest that can override this value; but I've not seen
-         * anything with that set.
-         */
-        host_phys_bits = eax & 0xff;
-    } else {
-        /* It's an odd 64 bit machine that doesn't have the leaf for
-         * physical address bits; fall back to 36 that's most older
-         * Intel.
-         */
-        host_phys_bits = 36;
-    }
-
-    return host_phys_bits;
-}
-
 static void x86_cpu_adjust_level(X86CPU *cpu, uint32_t *min, uint32_t value)
 {
     if (*min < value) {
@@ -6578,33 +6339,22 @@ static void x86_cpu_hyperv_realize(X86CPU *cpu)
 static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
 {
     CPUState *cs = CPU(dev);
+    CPUClass *cc = CPU_GET_CLASS(cs);
     X86CPU *cpu = X86_CPU(dev);
     X86CPUClass *xcc = X86_CPU_GET_CLASS(dev);
     CPUX86State *env = &cpu->env;
     Error *local_err = NULL;
     static bool ht_warned;
 
-    if (xcc->host_cpuid_required) {
-        if (!accel_uses_host_cpuid()) {
-            g_autofree char *name = x86_cpu_class_get_model_name(xcc);
-            error_setg(&local_err, "CPU model '%s' requires KVM", name);
-            goto out;
-        }
+    /* The accelerator realizefn needs to be called first. */
+    if (cc->accel_cpu) {
+        cc->accel_cpu->cpu_realizefn(cs, errp);
     }
 
-    if (cpu->max_features && accel_uses_host_cpuid()) {
-        if (enable_cpu_pm) {
-            host_cpuid(5, 0, &cpu->mwait.eax, &cpu->mwait.ebx,
-                       &cpu->mwait.ecx, &cpu->mwait.edx);
-            env->features[FEAT_1_ECX] |= CPUID_EXT_MONITOR;
-            if (kvm_enabled() && kvm_has_waitpkg()) {
-                env->features[FEAT_7_0_ECX] |= CPUID_7_0_ECX_WAITPKG;
-            }
-        }
-        if (kvm_enabled() && cpu->ucode_rev == 0) {
-            cpu->ucode_rev = kvm_arch_get_supported_msr_feature(kvm_state,
-                                                                MSR_IA32_UCODE_REV);
-        }
+    if (xcc->host_cpuid_required && !accel_uses_host_cpuid()) {
+        g_autofree char *name = x86_cpu_class_get_model_name(xcc);
+        error_setg(&local_err, "CPU model '%s' requires KVM or HVF", name);
+        goto out;
     }
 
     if (cpu->ucode_rev == 0) {
@@ -6656,39 +6406,7 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
      * consumer AMD devices but nothing else.
      */
     if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM) {
-        if (accel_uses_host_cpuid()) {
-            uint32_t host_phys_bits = x86_host_phys_bits();
-            static bool warned;
-
-            /* Print a warning if the user set it to a value that's not the
-             * host value.
-             */
-            if (cpu->phys_bits != host_phys_bits && cpu->phys_bits != 0 &&
-                !warned) {
-                warn_report("Host physical bits (%u)"
-                            " does not match phys-bits property (%u)",
-                            host_phys_bits, cpu->phys_bits);
-                warned = true;
-            }
-
-            if (cpu->host_phys_bits) {
-                /* The user asked for us to use the host physical bits */
-                cpu->phys_bits = host_phys_bits;
-                if (cpu->host_phys_bits_limit &&
-                    cpu->phys_bits > cpu->host_phys_bits_limit) {
-                    cpu->phys_bits = cpu->host_phys_bits_limit;
-                }
-            }
-
-            if (cpu->phys_bits &&
-                (cpu->phys_bits > TARGET_PHYS_ADDR_SPACE_BITS ||
-                cpu->phys_bits < 32)) {
-                error_setg(errp, "phys-bits should be between 32 and %u "
-                                 " (but is %u)",
-                                 TARGET_PHYS_ADDR_SPACE_BITS, cpu->phys_bits);
-                return;
-            }
-        } else {
+        if (!accel_uses_host_cpuid()) {
             if (cpu->phys_bits && cpu->phys_bits != TCG_PHYS_ADDR_BITS) {
                 error_setg(errp, "TCG only supports phys-bits=%u",
                                   TCG_PHYS_ADDR_BITS);
@@ -6696,8 +6414,8 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
             }
         }
         /* 0 means it was not explicitly set by the user (or by machine
-         * compat_props or by the host code above). In this case, the default
-         * is the value used by TCG (40).
+         * compat_props or by the host code in host-cpu.c).
+         * In this case, the default is the value used by TCG (40).
          */
         if (cpu->phys_bits == 0) {
             cpu->phys_bits = TCG_PHYS_ADDR_BITS;
@@ -6769,33 +6487,6 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
 
     mce_init(cpu);
 
-#ifndef CONFIG_USER_ONLY
-    if (tcg_enabled()) {
-        cpu->cpu_as_mem = g_new(MemoryRegion, 1);
-        cpu->cpu_as_root = g_new(MemoryRegion, 1);
-
-        /* Outer container... */
-        memory_region_init(cpu->cpu_as_root, OBJECT(cpu), "memory", ~0ull);
-        memory_region_set_enabled(cpu->cpu_as_root, true);
-
-        /* ... with two regions inside: normal system memory with low
-         * priority, and...
-         */
-        memory_region_init_alias(cpu->cpu_as_mem, OBJECT(cpu), "memory",
-                                 get_system_memory(), 0, ~0ull);
-        memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);
-        memory_region_set_enabled(cpu->cpu_as_mem, true);
-
-        cs->num_ases = 2;
-        cpu_address_space_init(cs, 0, "cpu-memory", cs->memory);
-        cpu_address_space_init(cs, 1, "cpu-smm", cpu->cpu_as_root);
-
-        /* ... SMRAM with higher priority, linked from /machine/smram.  */
-        cpu->machine_done.notify = x86_cpu_machine_done;
-        qemu_add_machine_init_done_notifier(&cpu->machine_done);
-    }
-#endif
-
     qemu_init_vcpu(cs);
 
     /*
@@ -6995,6 +6686,8 @@ static void x86_cpu_initfn(Object *obj)
 {
     X86CPU *cpu = X86_CPU(obj);
     X86CPUClass *xcc = X86_CPU_GET_CLASS(obj);
+    CPUClass *cc = CPU_CLASS(xcc);
+
     CPUX86State *env = &cpu->env;
 
     env->nr_dies = 1;
@@ -7042,6 +6735,11 @@ static void x86_cpu_initfn(Object *obj)
     if (xcc->model) {
         x86_cpu_load_model(cpu, xcc->model);
     }
+
+    /* if required, do the accelerator-specific cpu initialization */
+    if (cc->accel_cpu) {
+        cc->accel_cpu->cpu_instance_init(CPU(obj));
+    }
 }
 
 static int64_t x86_cpu_get_arch_id(CPUState *cs)
@@ -7299,11 +6997,6 @@ static void x86_cpu_common_class_init(ObjectClass *oc, void *data)
     cc->class_by_name = x86_cpu_class_by_name;
     cc->parse_features = x86_cpu_parse_featurestr;
     cc->has_work = x86_cpu_has_work;
-
-#ifdef CONFIG_TCG
-    tcg_cpu_common_class_init(cc);
-#endif /* CONFIG_TCG */
-
     cc->dump_state = x86_cpu_dump_state;
     cc->set_pc = x86_cpu_set_pc;
     cc->gdb_read_register = x86_cpu_gdb_read_register;
@@ -7414,9 +7107,6 @@ static void x86_cpu_register_types(void)
     }
     type_register_static(&max_x86_cpu_type_info);
     type_register_static(&x86_base_cpu_type_info);
-#if defined(CONFIG_KVM) || defined(CONFIG_HVF)
-    type_register_static(&host_x86_cpu_type_info);
-#endif
 }
 
 type_init(x86_cpu_register_types)
diff --git a/target/i386/host-cpu.c b/target/i386/host-cpu.c
new file mode 100644
index 0000000000..9cfe56ce41
--- /dev/null
+++ b/target/i386/host-cpu.c
@@ -0,0 +1,201 @@
+/*
+ * x86 host CPU functions, and "host" cpu type initialization
+ *
+ * Copyright 2021 SUSE LLC
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "host-cpu.h"
+#include "qapi/error.h"
+#include "sysemu/sysemu.h"
+
+/* Note: Only safe for use on x86(-64) hosts */
+static uint32_t host_cpu_phys_bits(void)
+{
+    uint32_t eax;
+    uint32_t host_phys_bits;
+
+    host_cpuid(0x80000000, 0, &eax, NULL, NULL, NULL);
+    if (eax >= 0x80000008) {
+        host_cpuid(0x80000008, 0, &eax, NULL, NULL, NULL);
+        /*
+         * Note: According to AMD doc 25481 rev 2.34 they have a field
+         * at 23:16 that can specify a maximum physical address bits for
+         * the guest that can override this value; but I've not seen
+         * anything with that set.
+         */
+        host_phys_bits = eax & 0xff;
+    } else {
+        /*
+         * It's an odd 64 bit machine that doesn't have the leaf for
+         * physical address bits; fall back to 36 that's most older
+         * Intel.
+         */
+        host_phys_bits = 36;
+    }
+
+    return host_phys_bits;
+}
+
+static void host_cpu_enable_cpu_pm(X86CPU *cpu)
+{
+    CPUX86State *env = &cpu->env;
+
+    host_cpuid(5, 0, &cpu->mwait.eax, &cpu->mwait.ebx,
+               &cpu->mwait.ecx, &cpu->mwait.edx);
+    env->features[FEAT_1_ECX] |= CPUID_EXT_MONITOR;
+}
+
+static uint32_t host_cpu_adjust_phys_bits(X86CPU *cpu, Error **errp)
+{
+    uint32_t host_phys_bits = host_cpu_phys_bits();
+    uint32_t phys_bits = cpu->phys_bits;
+    static bool warned;
+
+    /*
+     * Print a warning if the user set it to a value that's not the
+     * host value.
+     */
+    if (phys_bits != host_phys_bits && phys_bits != 0 &&
+        !warned) {
+        warn_report("Host physical bits (%u)"
+                    " does not match phys-bits property (%u)",
+                    host_phys_bits, phys_bits);
+        warned = true;
+    }
+
+    if (cpu->host_phys_bits) {
+        /* The user asked for us to use the host physical bits */
+        phys_bits = host_phys_bits;
+        if (cpu->host_phys_bits_limit &&
+            phys_bits > cpu->host_phys_bits_limit) {
+            phys_bits = cpu->host_phys_bits_limit;
+        }
+    }
+
+    if (phys_bits &&
+        (phys_bits > TARGET_PHYS_ADDR_SPACE_BITS ||
+         phys_bits < 32)) {
+        error_setg(errp, "phys-bits should be between 32 and %u "
+                   " (but is %u)",
+                   TARGET_PHYS_ADDR_SPACE_BITS, phys_bits);
+    }
+
+    return phys_bits;
+}
+
+void host_cpu_realizefn(CPUState *cs, Error **errp)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    if (cpu->max_features && enable_cpu_pm) {
+        host_cpu_enable_cpu_pm(cpu);
+    }
+    if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM) {
+        cpu->phys_bits = host_cpu_adjust_phys_bits(cpu, errp);
+    }
+}
+
+#define CPUID_MODEL_ID_SZ 48
+/**
+ * cpu_x86_fill_model_id:
+ * Get CPUID model ID string from host CPU.
+ *
+ * @str should have at least CPUID_MODEL_ID_SZ bytes
+ *
+ * The function does NOT add a null terminator to the string
+ * automatically.
+ */
+static int host_cpu_fill_model_id(char *str)
+{
+    uint32_t eax = 0, ebx = 0, ecx = 0, edx = 0;
+    int i;
+
+    for (i = 0; i < 3; i++) {
+        host_cpuid(0x80000002 + i, 0, &eax, &ebx, &ecx, &edx);
+        memcpy(str + i * 16 +  0, &eax, 4);
+        memcpy(str + i * 16 +  4, &ebx, 4);
+        memcpy(str + i * 16 +  8, &ecx, 4);
+        memcpy(str + i * 16 + 12, &edx, 4);
+    }
+    return 0;
+}
+
+void host_cpu_vendor_fms(char *vendor, int *family, int *model, int *stepping)
+{
+    uint32_t eax, ebx, ecx, edx;
+
+    host_cpuid(0x0, 0, &eax, &ebx, &ecx, &edx);
+    x86_cpu_vendor_words2str(vendor, ebx, edx, ecx);
+
+    host_cpuid(0x1, 0, &eax, &ebx, &ecx, &edx);
+    if (family) {
+        *family = ((eax >> 8) & 0x0F) + ((eax >> 20) & 0xFF);
+    }
+    if (model) {
+        *model = ((eax >> 4) & 0x0F) | ((eax & 0xF0000) >> 12);
+    }
+    if (stepping) {
+        *stepping = eax & 0x0F;
+    }
+}
+
+void host_cpu_instance_init(X86CPU *cpu)
+{
+    uint32_t ebx = 0, ecx = 0, edx = 0;
+    char vendor[CPUID_VENDOR_SZ + 1];
+
+    host_cpuid(0, 0, NULL, &ebx, &ecx, &edx);
+    x86_cpu_vendor_words2str(vendor, ebx, edx, ecx);
+
+    object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
+}
+
+void host_cpu_max_instance_init(X86CPU *cpu)
+{
+    char vendor[CPUID_VENDOR_SZ + 1] = { 0 };
+    char model_id[CPUID_MODEL_ID_SZ + 1] = { 0 };
+    int family, model, stepping;
+
+    /* Use max host physical address bits if -cpu max option is applied */
+    object_property_set_bool(OBJECT(cpu), "host-phys-bits", true, &error_abort);
+
+    host_cpu_vendor_fms(vendor, &family, &model, &stepping);
+    host_cpu_fill_model_id(model_id);
+
+    object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
+    object_property_set_int(OBJECT(cpu), "family", family, &error_abort);
+    object_property_set_int(OBJECT(cpu), "model", model, &error_abort);
+    object_property_set_int(OBJECT(cpu), "stepping", stepping,
+                            &error_abort);
+    object_property_set_str(OBJECT(cpu), "model-id", model_id,
+                            &error_abort);
+}
+
+static void host_cpu_class_init(ObjectClass *oc, void *data)
+{
+    X86CPUClass *xcc = X86_CPU_CLASS(oc);
+
+    xcc->host_cpuid_required = true;
+    xcc->ordering = 8;
+    xcc->model_description =
+        g_strdup_printf("processor with all supported host features ");
+}
+
+static const TypeInfo host_cpu_type_info = {
+    .name = X86_CPU_TYPE_NAME("host"),
+    .parent = X86_CPU_TYPE_NAME("max"),
+    .class_init = host_cpu_class_init,
+};
+
+static void host_cpu_type_init(void)
+{
+    type_register_static(&host_cpu_type_info);
+}
+
+type_init(host_cpu_type_init);
diff --git a/target/i386/hvf/hvf-cpu.c b/target/i386/hvf/hvf-cpu.c
new file mode 100644
index 0000000000..8fbc423888
--- /dev/null
+++ b/target/i386/hvf/hvf-cpu.c
@@ -0,0 +1,68 @@
+/*
+ * x86 HVF CPU type initialization
+ *
+ * Copyright 2021 SUSE LLC
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "host-cpu.h"
+#include "qapi/error.h"
+#include "sysemu/sysemu.h"
+#include "hw/boards.h"
+#include "sysemu/hvf.h"
+#include "hw/core/accel-cpu.h"
+
+static void hvf_cpu_max_instance_init(X86CPU *cpu)
+{
+    CPUX86State *env = &cpu->env;
+
+    host_cpu_max_instance_init(cpu);
+
+    env->cpuid_min_level =
+        hvf_get_supported_cpuid(0x0, 0, R_EAX);
+    env->cpuid_min_xlevel =
+        hvf_get_supported_cpuid(0x80000000, 0, R_EAX);
+    env->cpuid_min_xlevel2 =
+        hvf_get_supported_cpuid(0xC0000000, 0, R_EAX);
+}
+
+static void hvf_cpu_instance_init(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+
+    host_cpu_instance_init(cpu);
+
+    /* Special cases not set in the X86CPUDefinition structs: */
+    /* TODO: in-kernel irqchip for hvf */
+
+    if (cpu->max_features) {
+        hvf_cpu_max_instance_init(cpu);
+    }
+}
+
+static void hvf_cpu_accel_class_init(ObjectClass *oc, void *data)
+{
+    AccelCPUClass *acc = ACCEL_CPU_CLASS(oc);
+
+    acc->cpu_realizefn = host_cpu_realizefn;
+    acc->cpu_instance_init = hvf_cpu_instance_init;
+}
+
+static const TypeInfo hvf_cpu_accel_type_info = {
+    .name = ACCEL_CPU_NAME("hvf"),
+
+    .parent = TYPE_ACCEL_CPU,
+    .class_init = hvf_cpu_accel_class_init,
+    .abstract = true,
+};
+
+static void hvf_cpu_accel_register_types(void)
+{
+    type_register_static(&hvf_cpu_accel_type_info);
+}
+
+type_init(hvf_cpu_accel_register_types);
diff --git a/target/i386/kvm/kvm-cpu.c b/target/i386/kvm/kvm-cpu.c
new file mode 100644
index 0000000000..c23bbe6c50
--- /dev/null
+++ b/target/i386/kvm/kvm-cpu.c
@@ -0,0 +1,151 @@
+/*
+ * x86 KVM CPU type initialization
+ *
+ * Copyright 2021 SUSE LLC
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "host-cpu.h"
+#include "kvm-cpu.h"
+#include "qapi/error.h"
+#include "sysemu/sysemu.h"
+#include "hw/boards.h"
+
+#include "kvm_i386.h"
+#include "hw/core/accel-cpu.h"
+
+static void kvm_cpu_realizefn(CPUState *cs, Error **errp)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    /*
+     * The realize order is important, since x86_cpu_realize() checks if
+     * nothing else has been set by the user (or by accelerators) in
+     * cpu->ucode_rev and cpu->phys_bits.
+     *
+     * realize order:
+     * kvm_cpu -> host_cpu -> x86_cpu
+     */
+    if (cpu->max_features) {
+        if (enable_cpu_pm && kvm_has_waitpkg()) {
+            env->features[FEAT_7_0_ECX] |= CPUID_7_0_ECX_WAITPKG;
+        }
+        if (cpu->ucode_rev == 0) {
+            cpu->ucode_rev =
+                kvm_arch_get_supported_msr_feature(kvm_state,
+                                                   MSR_IA32_UCODE_REV);
+        }
+    }
+    host_cpu_realizefn(cs, errp);
+}
+
+/*
+ * KVM-specific features that are automatically added/removed
+ * from all CPU models when KVM is enabled.
+ */
+static PropValue kvm_default_props[] = {
+    { "kvmclock", "on" },
+    { "kvm-nopiodelay", "on" },
+    { "kvm-asyncpf", "on" },
+    { "kvm-steal-time", "on" },
+    { "kvm-pv-eoi", "on" },
+    { "kvmclock-stable-bit", "on" },
+    { "x2apic", "on" },
+    { "kvm-msi-ext-dest-id", "off" },
+    { "acpi", "off" },
+    { "monitor", "off" },
+    { "svm", "off" },
+    { NULL, NULL },
+};
+
+void x86_cpu_change_kvm_default(const char *prop, const char *value)
+{
+    PropValue *pv;
+    for (pv = kvm_default_props; pv->prop; pv++) {
+        if (!strcmp(pv->prop, prop)) {
+            pv->value = value;
+            break;
+        }
+    }
+
+    /*
+     * It is valid to call this function only for properties that
+     * are already present in the kvm_default_props table.
+     */
+    assert(pv->prop);
+}
+
+static bool lmce_supported(void)
+{
+    uint64_t mce_cap = 0;
+
+    if (kvm_ioctl(kvm_state, KVM_X86_GET_MCE_CAP_SUPPORTED, &mce_cap) < 0) {
+        return false;
+    }
+    return !!(mce_cap & MCG_LMCE_P);
+}
+
+static void kvm_cpu_max_instance_init(X86CPU *cpu)
+{
+    CPUX86State *env = &cpu->env;
+    KVMState *s = kvm_state;
+
+    host_cpu_max_instance_init(cpu);
+
+    if (lmce_supported()) {
+        object_property_set_bool(OBJECT(cpu), "lmce", true, &error_abort);
+    }
+
+    env->cpuid_min_level =
+        kvm_arch_get_supported_cpuid(s, 0x0, 0, R_EAX);
+    env->cpuid_min_xlevel =
+        kvm_arch_get_supported_cpuid(s, 0x80000000, 0, R_EAX);
+    env->cpuid_min_xlevel2 =
+        kvm_arch_get_supported_cpuid(s, 0xC0000000, 0, R_EAX);
+}
+
+static void kvm_cpu_instance_init(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+
+    host_cpu_instance_init(cpu);
+
+    if (!kvm_irqchip_in_kernel()) {
+        x86_cpu_change_kvm_default("x2apic", "off");
+    } else if (kvm_irqchip_is_split() && kvm_enable_x2apic()) {
+        x86_cpu_change_kvm_default("kvm-msi-ext-dest-id", "on");
+    }
+
+    /* Special cases not set in the X86CPUDefinition structs: */
+
+    x86_cpu_apply_props(cpu, kvm_default_props);
+
+    if (cpu->max_features) {
+        kvm_cpu_max_instance_init(cpu);
+    }
+}
+
+static void kvm_cpu_accel_class_init(ObjectClass *oc, void *data)
+{
+    AccelCPUClass *acc = ACCEL_CPU_CLASS(oc);
+
+    acc->cpu_realizefn = kvm_cpu_realizefn;
+    acc->cpu_instance_init = kvm_cpu_instance_init;
+}
+static const TypeInfo kvm_cpu_accel_type_info = {
+    .name = ACCEL_CPU_NAME("kvm"),
+
+    .parent = TYPE_ACCEL_CPU,
+    .class_init = kvm_cpu_accel_class_init,
+    .abstract = true,
+};
+static void kvm_cpu_accel_register_types(void)
+{
+    type_register_static(&kvm_cpu_accel_type_info);
+}
+type_init(kvm_cpu_accel_register_types);
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index e97f841757..5ac35d2801 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -22,6 +22,7 @@
 #include "standard-headers/asm-x86/kvm_para.h"
 
 #include "cpu.h"
+#include "host-cpu.h"
 #include "sysemu/sysemu.h"
 #include "sysemu/hw_accel.h"
 #include "sysemu/kvm_int.h"
@@ -287,7 +288,7 @@ static bool host_tsx_broken(void)
     int family, model, stepping;\
     char vendor[CPUID_VENDOR_SZ + 1];
 
-    host_vendor_fms(vendor, &family, &model, &stepping);
+    host_cpu_vendor_fms(vendor, &family, &model, &stepping);
 
     /* Check if we are running on a Haswell host known to have broken TSX */
     return !strcmp(vendor, CPUID_VENDOR_INTEL) &&
diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c
index 1e125d2175..1d3d6d1c6a 100644
--- a/target/i386/tcg/tcg-cpu.c
+++ b/target/i386/tcg/tcg-cpu.c
@@ -19,13 +19,14 @@
 
 #include "qemu/osdep.h"
 #include "cpu.h"
-#include "tcg-cpu.h"
-#include "exec/exec-all.h"
-#include "sysemu/runstate.h"
 #include "helper-tcg.h"
+#include "qemu/accel.h"
+#include "hw/core/accel-cpu.h"
 
-#if !defined(CONFIG_USER_ONLY)
-#include "hw/i386/apic.h"
+#ifndef CONFIG_USER_ONLY
+#include "sysemu/sysemu.h"
+#include "qemu/units.h"
+#include "exec/address-spaces.h"
 #endif
 
 /* Frob eflags into and out of the CPU temporary format.  */
@@ -72,7 +73,107 @@ static struct TCGCPUOps x86_tcg_ops = {
 #endif /* !CONFIG_USER_ONLY */
 };
 
-void tcg_cpu_common_class_init(CPUClass *cc)
+static void tcg_cpu_class_init(CPUClass *cc)
 {
     cc->tcg_ops = &x86_tcg_ops;
 }
+
+#ifndef CONFIG_USER_ONLY
+
+static void x86_cpu_machine_done(Notifier *n, void *unused)
+{
+    X86CPU *cpu = container_of(n, X86CPU, machine_done);
+    MemoryRegion *smram =
+        (MemoryRegion *) object_resolve_path("/machine/smram", NULL);
+
+    if (smram) {
+        cpu->smram = g_new(MemoryRegion, 1);
+        memory_region_init_alias(cpu->smram, OBJECT(cpu), "smram",
+                                 smram, 0, 4 * GiB);
+        memory_region_set_enabled(cpu->smram, true);
+        memory_region_add_subregion_overlap(cpu->cpu_as_root, 0,
+                                            cpu->smram, 1);
+    }
+}
+
+static void tcg_cpu_realizefn(CPUState *cs, Error **errp)
+{
+    X86CPU *cpu = X86_CPU(cs);
+
+    /*
+     * The realize order is important, since x86_cpu_realize() checks if
+     * nothing else has been set by the user (or by accelerators) in
+     * cpu->ucode_rev and cpu->phys_bits, and the memory regions
+     * initialized here are needed for the vcpu initialization.
+     *
+     * realize order:
+     * tcg_cpu -> host_cpu -> x86_cpu
+     */
+    cpu->cpu_as_mem = g_new(MemoryRegion, 1);
+    cpu->cpu_as_root = g_new(MemoryRegion, 1);
+
+    /* Outer container... */
+    memory_region_init(cpu->cpu_as_root, OBJECT(cpu), "memory", ~0ull);
+    memory_region_set_enabled(cpu->cpu_as_root, true);
+
+    /*
+     * ... with two regions inside: normal system memory with low
+     * priority, and...
+     */
+    memory_region_init_alias(cpu->cpu_as_mem, OBJECT(cpu), "memory",
+                             get_system_memory(), 0, ~0ull);
+    memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);
+    memory_region_set_enabled(cpu->cpu_as_mem, true);
+
+    cs->num_ases = 2;
+    cpu_address_space_init(cs, 0, "cpu-memory", cs->memory);
+    cpu_address_space_init(cs, 1, "cpu-smm", cpu->cpu_as_root);
+
+    /* ... SMRAM with higher priority, linked from /machine/smram.  */
+    cpu->machine_done.notify = x86_cpu_machine_done;
+    qemu_add_machine_init_done_notifier(&cpu->machine_done);
+}
+
+#else /* CONFIG_USER_ONLY */
+
+static void tcg_cpu_realizefn(CPUState *cs, Error **errp)
+{
+}
+
+#endif /* !CONFIG_USER_ONLY */
+
+/*
+ * TCG-specific defaults that override all CPU models when using TCG
+ */
+static PropValue tcg_default_props[] = {
+    { "vme", "off" },
+    { NULL, NULL },
+};
+
+static void tcg_cpu_instance_init(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    /* Special cases not set in the X86CPUDefinition structs: */
+    x86_cpu_apply_props(cpu, tcg_default_props);
+}
+
+static void tcg_cpu_accel_class_init(ObjectClass *oc, void *data)
+{
+    AccelCPUClass *acc = ACCEL_CPU_CLASS(oc);
+
+    acc->cpu_realizefn = tcg_cpu_realizefn;
+    acc->cpu_class_init = tcg_cpu_class_init;
+    acc->cpu_instance_init = tcg_cpu_instance_init;
+}
+static const TypeInfo tcg_cpu_accel_type_info = {
+    .name = ACCEL_CPU_NAME("tcg"),
+
+    .parent = TYPE_ACCEL_CPU,
+    .class_init = tcg_cpu_accel_class_init,
+    .abstract = true,
+};
+static void tcg_cpu_accel_register_types(void)
+{
+    type_register_static(&tcg_cpu_accel_type_info);
+}
+type_init(tcg_cpu_accel_register_types);
diff --git a/MAINTAINERS b/MAINTAINERS
index e6f1eca30f..54140d0725 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -341,7 +341,7 @@ M: Paolo Bonzini <pbonzini@redhat.com>
 M: Richard Henderson <richard.henderson@linaro.org>
 M: Eduardo Habkost <ehabkost@redhat.com>
 S: Maintained
-F: target/i386/
+F: target/i386/tcg/
 F: tests/tcg/i386/
 F: tests/tcg/x86_64/
 F: hw/i386/
diff --git a/target/i386/hvf/meson.build b/target/i386/hvf/meson.build
index e9eb5a5da8..d253d5fd10 100644
--- a/target/i386/hvf/meson.build
+++ b/target/i386/hvf/meson.build
@@ -10,4 +10,5 @@ i386_softmmu_ss.add(when: [hvf, 'CONFIG_HVF'], if_true: files(
   'x86_mmu.c',
   'x86_task.c',
   'x86hvf.c',
+  'hvf-cpu.c',
 ))
diff --git a/target/i386/kvm/meson.build b/target/i386/kvm/meson.build
index 1d66559187..0a533411ca 100644
--- a/target/i386/kvm/meson.build
+++ b/target/i386/kvm/meson.build
@@ -1,3 +1,8 @@
 i386_ss.add(when: 'CONFIG_KVM', if_false: files('kvm-stub.c'))
-i386_softmmu_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
+
+i386_softmmu_ss.add(when: 'CONFIG_KVM', if_true: files(
+  'kvm.c',
+  'kvm-cpu.c',
+))
+
 i386_softmmu_ss.add(when: 'CONFIG_HYPERV', if_true: files('hyperv.c'), if_false: files('hyperv-stub.c'))
diff --git a/target/i386/meson.build b/target/i386/meson.build
index c4bf20b319..fd24479590 100644
--- a/target/i386/meson.build
+++ b/target/i386/meson.build
@@ -6,7 +6,11 @@ i386_ss.add(files(
   'xsave_helper.c',
   'cpu-dump.c',
 ))
-i386_ss.add(when: 'CONFIG_SEV', if_true: files('sev.c'), if_false: files('sev-stub.c'))
+i386_ss.add(when: 'CONFIG_SEV', if_true: files('host-cpu.c', 'sev.c'), if_false: files('sev-stub.c'))
+
+# x86 cpu type
+i386_ss.add(when: 'CONFIG_KVM', if_true: files('host-cpu.c'))
+i386_ss.add(when: 'CONFIG_HVF', if_true: files('host-cpu.c'))
 
 i386_softmmu_ss = ss.source_set()
 i386_softmmu_ss.add(files(
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:30   ` Alex Bennée
  2021-02-12 12:36 ` [RFC v18 03/15] accel: introduce new accessor functions Claudio Fontana
                   ` (14 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

move the call to accel_cpu->cpu_realizefn to the general
cpu_exec_realizefn from target/i386, so it does not need to be
called for every target explicitly as we enable more targets.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 cpu.c             |  6 ++++++
 target/i386/cpu.c | 20 +++++++-------------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/cpu.c b/cpu.c
index bfbe5a66f9..ba5d272c1e 100644
--- a/cpu.c
+++ b/cpu.c
@@ -36,6 +36,7 @@
 #include "sysemu/replay.h"
 #include "exec/translate-all.h"
 #include "exec/log.h"
+#include "hw/core/accel-cpu.h"
 
 uintptr_t qemu_host_page_size;
 intptr_t qemu_host_page_mask;
@@ -130,6 +131,11 @@ void cpu_exec_realizefn(CPUState *cpu, Error **errp)
 
     cpu_list_add(cpu);
 
+    if (cc->accel_cpu) {
+        /* NB: errp parameter is unused currently */
+        cc->accel_cpu->cpu_realizefn(cpu, errp);
+    }
+
 #ifdef CONFIG_TCG
     /* NB: errp parameter is unused currently */
     if (tcg_enabled()) {
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index b034571869..a81a1edcd4 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -6339,16 +6339,19 @@ static void x86_cpu_hyperv_realize(X86CPU *cpu)
 static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
 {
     CPUState *cs = CPU(dev);
-    CPUClass *cc = CPU_GET_CLASS(cs);
     X86CPU *cpu = X86_CPU(dev);
     X86CPUClass *xcc = X86_CPU_GET_CLASS(dev);
     CPUX86State *env = &cpu->env;
     Error *local_err = NULL;
     static bool ht_warned;
 
-    /* The accelerator realizefn needs to be called first. */
-    if (cc->accel_cpu) {
-        cc->accel_cpu->cpu_realizefn(cs, errp);
+    /* Process Hyper-V enlightenments */
+    x86_cpu_hyperv_realize(cpu);
+
+    cpu_exec_realizefn(cs, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
+        return;
     }
 
     if (xcc->host_cpuid_required && !accel_uses_host_cpuid()) {
@@ -6464,15 +6467,6 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
         env->cache_info_amd.l3_cache = &legacy_l3_cache;
     }
 
-    /* Process Hyper-V enlightenments */
-    x86_cpu_hyperv_realize(cpu);
-
-    cpu_exec_realizefn(cs, &local_err);
-    if (local_err != NULL) {
-        error_propagate(errp, local_err);
-        return;
-    }
-
 #ifndef CONFIG_USER_ONLY
     MachineState *ms = MACHINE(qdev_get_machine());
     qemu_register_reset(x86_cpu_machine_reset_cb, cpu);
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 03/15] accel: introduce new accessor functions
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:34   ` Alex Bennée
  2021-02-12 12:36 ` [RFC v18 04/15] target/i386: fix host_cpu_adjust_phys_bits error handling Claudio Fontana
                   ` (13 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

avoid open coding the accesses to cpu->accel_cpu interfaces,
and instead introduce:

accel_cpu_instance_init,
accel_cpu_realizefn

to be used by the targets/ initfn code,
and by cpu_exec_realizefn respectively.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 include/qemu/accel.h | 13 +++++++++++++
 accel/accel-common.c | 19 +++++++++++++++++++
 cpu.c                |  6 +-----
 target/i386/cpu.c    |  9 ++-------
 4 files changed, 35 insertions(+), 12 deletions(-)

diff --git a/include/qemu/accel.h b/include/qemu/accel.h
index b9d6d69eb8..da0c8ab523 100644
--- a/include/qemu/accel.h
+++ b/include/qemu/accel.h
@@ -78,4 +78,17 @@ int accel_init_machine(AccelState *accel, MachineState *ms);
 void accel_setup_post(MachineState *ms);
 #endif /* !CONFIG_USER_ONLY */
 
+/**
+ * accel_cpu_instance_init:
+ * @cpu: The CPU that needs to do accel-specific object initializations.
+ */
+void accel_cpu_instance_init(CPUState *cpu);
+
+/**
+ * accel_cpu_realizefn:
+ * @cpu: The CPU that needs to call accel-specific cpu realization.
+ * @errp: currently unused.
+ */
+void accel_cpu_realizefn(CPUState *cpu, Error **errp);
+
 #endif /* QEMU_ACCEL_H */
diff --git a/accel/accel-common.c b/accel/accel-common.c
index 9901b0531c..0f6fb4fb66 100644
--- a/accel/accel-common.c
+++ b/accel/accel-common.c
@@ -89,6 +89,25 @@ void accel_init_interfaces(AccelClass *ac)
     accel_init_cpu_interfaces(ac);
 }
 
+void accel_cpu_instance_init(CPUState *cpu)
+{
+    CPUClass *cc = CPU_GET_CLASS(cpu);
+
+    if (cc->accel_cpu && cc->accel_cpu->cpu_instance_init) {
+        cc->accel_cpu->cpu_instance_init(cpu);
+    }
+}
+
+void accel_cpu_realizefn(CPUState *cpu, Error **errp)
+{
+    CPUClass *cc = CPU_GET_CLASS(cpu);
+
+    if (cc->accel_cpu && cc->accel_cpu->cpu_realizefn) {
+        /* NB: errp parameter is unused currently */
+        cc->accel_cpu->cpu_realizefn(cpu, errp);
+    }
+}
+
 static const TypeInfo accel_cpu_type = {
     .name = TYPE_ACCEL_CPU,
     .parent = TYPE_OBJECT,
diff --git a/cpu.c b/cpu.c
index ba5d272c1e..25e6fbfa2c 100644
--- a/cpu.c
+++ b/cpu.c
@@ -130,11 +130,7 @@ void cpu_exec_realizefn(CPUState *cpu, Error **errp)
     CPUClass *cc = CPU_GET_CLASS(cpu);
 
     cpu_list_add(cpu);
-
-    if (cc->accel_cpu) {
-        /* NB: errp parameter is unused currently */
-        cc->accel_cpu->cpu_realizefn(cpu, errp);
-    }
+    accel_cpu_realizefn(cpu, errp);
 
 #ifdef CONFIG_TCG
     /* NB: errp parameter is unused currently */
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index a81a1edcd4..8d2e1c3136 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -28,7 +28,6 @@
 #include "sysemu/kvm.h"
 #include "sysemu/reset.h"
 #include "sysemu/hvf.h"
-#include "hw/core/accel-cpu.h"
 #include "sysemu/xen.h"
 #include "sysemu/whpx.h"
 #include "kvm/kvm_i386.h"
@@ -6680,8 +6679,6 @@ static void x86_cpu_initfn(Object *obj)
 {
     X86CPU *cpu = X86_CPU(obj);
     X86CPUClass *xcc = X86_CPU_GET_CLASS(obj);
-    CPUClass *cc = CPU_CLASS(xcc);
-
     CPUX86State *env = &cpu->env;
 
     env->nr_dies = 1;
@@ -6730,10 +6727,8 @@ static void x86_cpu_initfn(Object *obj)
         x86_cpu_load_model(cpu, xcc->model);
     }
 
-    /* if required, do the accelerator-specific cpu initialization */
-    if (cc->accel_cpu) {
-        cc->accel_cpu->cpu_instance_init(CPU(obj));
-    }
+    /* if required, do accelerator-specific cpu initializations */
+    accel_cpu_instance_init(CPU(obj));
 }
 
 static int64_t x86_cpu_get_arch_id(CPUState *cs)
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 04/15] target/i386: fix host_cpu_adjust_phys_bits error handling
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (2 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 03/15] accel: introduce new accessor functions Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 05/15] accel-cpu: make cpu_realizefn return a bool Claudio Fontana
                   ` (12 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

move the check for phys_bits outside of host_cpu_adjust_phys_bits,
because otherwise it is impossible to return an error condition
explicitly.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/i386/host-cpu.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/target/i386/host-cpu.c b/target/i386/host-cpu.c
index 9cfe56ce41..d07d41c34c 100644
--- a/target/i386/host-cpu.c
+++ b/target/i386/host-cpu.c
@@ -50,7 +50,7 @@ static void host_cpu_enable_cpu_pm(X86CPU *cpu)
     env->features[FEAT_1_ECX] |= CPUID_EXT_MONITOR;
 }
 
-static uint32_t host_cpu_adjust_phys_bits(X86CPU *cpu, Error **errp)
+static uint32_t host_cpu_adjust_phys_bits(X86CPU *cpu)
 {
     uint32_t host_phys_bits = host_cpu_phys_bits();
     uint32_t phys_bits = cpu->phys_bits;
@@ -77,14 +77,6 @@ static uint32_t host_cpu_adjust_phys_bits(X86CPU *cpu, Error **errp)
         }
     }
 
-    if (phys_bits &&
-        (phys_bits > TARGET_PHYS_ADDR_SPACE_BITS ||
-         phys_bits < 32)) {
-        error_setg(errp, "phys-bits should be between 32 and %u "
-                   " (but is %u)",
-                   TARGET_PHYS_ADDR_SPACE_BITS, phys_bits);
-    }
-
     return phys_bits;
 }
 
@@ -97,7 +89,17 @@ void host_cpu_realizefn(CPUState *cs, Error **errp)
         host_cpu_enable_cpu_pm(cpu);
     }
     if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM) {
-        cpu->phys_bits = host_cpu_adjust_phys_bits(cpu, errp);
+        uint32_t phys_bits = host_cpu_adjust_phys_bits(cpu);
+
+        if (phys_bits &&
+            (phys_bits > TARGET_PHYS_ADDR_SPACE_BITS ||
+             phys_bits < 32)) {
+            error_setg(errp, "phys-bits should be between 32 and %u "
+                       " (but is %u)",
+                       TARGET_PHYS_ADDR_SPACE_BITS, phys_bits);
+            return;
+        }
+        cpu->phys_bits = phys_bits;
     }
 }
 
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 05/15] accel-cpu: make cpu_realizefn return a bool
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (3 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 04/15] target/i386: fix host_cpu_adjust_phys_bits error handling Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 06/15] meson: add target_user_arch Claudio Fontana
                   ` (11 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

overall, all devices' realize functions take an Error **errp, but return void.

hw/core/qdev.c code, which realizes devices, therefore does:

local_err = NULL;
dc->realize(dev, &local_err);
if (local_err != NULL) {
    goto fail;
}

However, we can improve at least accel_cpu to return a meaningful bool value.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
---
 include/hw/core/accel-cpu.h | 2 +-
 include/qemu/accel.h        | 2 +-
 target/i386/host-cpu.h      | 2 +-
 accel/accel-common.c        | 6 +++---
 cpu.c                       | 5 +++--
 target/i386/host-cpu.c      | 5 +++--
 target/i386/kvm/kvm-cpu.c   | 4 ++--
 target/i386/tcg/tcg-cpu.c   | 6 ++++--
 8 files changed, 18 insertions(+), 14 deletions(-)

diff --git a/include/hw/core/accel-cpu.h b/include/hw/core/accel-cpu.h
index 24a6697412..5dbfd79955 100644
--- a/include/hw/core/accel-cpu.h
+++ b/include/hw/core/accel-cpu.h
@@ -32,7 +32,7 @@ typedef struct AccelCPUClass {
 
     void (*cpu_class_init)(CPUClass *cc);
     void (*cpu_instance_init)(CPUState *cpu);
-    void (*cpu_realizefn)(CPUState *cpu, Error **errp);
+    bool (*cpu_realizefn)(CPUState *cpu, Error **errp);
 } AccelCPUClass;
 
 #endif /* ACCEL_CPU_H */
diff --git a/include/qemu/accel.h b/include/qemu/accel.h
index da0c8ab523..4f4c283f6f 100644
--- a/include/qemu/accel.h
+++ b/include/qemu/accel.h
@@ -89,6 +89,6 @@ void accel_cpu_instance_init(CPUState *cpu);
  * @cpu: The CPU that needs to call accel-specific cpu realization.
  * @errp: currently unused.
  */
-void accel_cpu_realizefn(CPUState *cpu, Error **errp);
+bool accel_cpu_realizefn(CPUState *cpu, Error **errp);
 
 #endif /* QEMU_ACCEL_H */
diff --git a/target/i386/host-cpu.h b/target/i386/host-cpu.h
index b47bc0943f..6a9bc918ba 100644
--- a/target/i386/host-cpu.h
+++ b/target/i386/host-cpu.h
@@ -12,7 +12,7 @@
 
 void host_cpu_instance_init(X86CPU *cpu);
 void host_cpu_max_instance_init(X86CPU *cpu);
-void host_cpu_realizefn(CPUState *cs, Error **errp);
+bool host_cpu_realizefn(CPUState *cs, Error **errp);
 
 void host_cpu_vendor_fms(char *vendor, int *family, int *model, int *stepping);
 
diff --git a/accel/accel-common.c b/accel/accel-common.c
index 0f6fb4fb66..d77c09d7b5 100644
--- a/accel/accel-common.c
+++ b/accel/accel-common.c
@@ -98,14 +98,14 @@ void accel_cpu_instance_init(CPUState *cpu)
     }
 }
 
-void accel_cpu_realizefn(CPUState *cpu, Error **errp)
+bool accel_cpu_realizefn(CPUState *cpu, Error **errp)
 {
     CPUClass *cc = CPU_GET_CLASS(cpu);
 
     if (cc->accel_cpu && cc->accel_cpu->cpu_realizefn) {
-        /* NB: errp parameter is unused currently */
-        cc->accel_cpu->cpu_realizefn(cpu, errp);
+        return cc->accel_cpu->cpu_realizefn(cpu, errp);
     }
+    return true;
 }
 
 static const TypeInfo accel_cpu_type = {
diff --git a/cpu.c b/cpu.c
index 25e6fbfa2c..34a0484bf4 100644
--- a/cpu.c
+++ b/cpu.c
@@ -130,8 +130,9 @@ void cpu_exec_realizefn(CPUState *cpu, Error **errp)
     CPUClass *cc = CPU_GET_CLASS(cpu);
 
     cpu_list_add(cpu);
-    accel_cpu_realizefn(cpu, errp);
-
+    if (!accel_cpu_realizefn(cpu, errp)) {
+        return;
+    }
 #ifdef CONFIG_TCG
     /* NB: errp parameter is unused currently */
     if (tcg_enabled()) {
diff --git a/target/i386/host-cpu.c b/target/i386/host-cpu.c
index d07d41c34c..4ea9e354ea 100644
--- a/target/i386/host-cpu.c
+++ b/target/i386/host-cpu.c
@@ -80,7 +80,7 @@ static uint32_t host_cpu_adjust_phys_bits(X86CPU *cpu)
     return phys_bits;
 }
 
-void host_cpu_realizefn(CPUState *cs, Error **errp)
+bool host_cpu_realizefn(CPUState *cs, Error **errp)
 {
     X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
@@ -97,10 +97,11 @@ void host_cpu_realizefn(CPUState *cs, Error **errp)
             error_setg(errp, "phys-bits should be between 32 and %u "
                        " (but is %u)",
                        TARGET_PHYS_ADDR_SPACE_BITS, phys_bits);
-            return;
+            return false;
         }
         cpu->phys_bits = phys_bits;
     }
+    return true;
 }
 
 #define CPUID_MODEL_ID_SZ 48
diff --git a/target/i386/kvm/kvm-cpu.c b/target/i386/kvm/kvm-cpu.c
index c23bbe6c50..c660ad4293 100644
--- a/target/i386/kvm/kvm-cpu.c
+++ b/target/i386/kvm/kvm-cpu.c
@@ -18,7 +18,7 @@
 #include "kvm_i386.h"
 #include "hw/core/accel-cpu.h"
 
-static void kvm_cpu_realizefn(CPUState *cs, Error **errp)
+static bool kvm_cpu_realizefn(CPUState *cs, Error **errp)
 {
     X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
@@ -41,7 +41,7 @@ static void kvm_cpu_realizefn(CPUState *cs, Error **errp)
                                                    MSR_IA32_UCODE_REV);
         }
     }
-    host_cpu_realizefn(cs, errp);
+    return host_cpu_realizefn(cs, errp);
 }
 
 /*
diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c
index 1d3d6d1c6a..23e1f5f0c3 100644
--- a/target/i386/tcg/tcg-cpu.c
+++ b/target/i386/tcg/tcg-cpu.c
@@ -96,7 +96,7 @@ static void x86_cpu_machine_done(Notifier *n, void *unused)
     }
 }
 
-static void tcg_cpu_realizefn(CPUState *cs, Error **errp)
+static bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
 {
     X86CPU *cpu = X86_CPU(cs);
 
@@ -132,12 +132,14 @@ static void tcg_cpu_realizefn(CPUState *cs, Error **errp)
     /* ... SMRAM with higher priority, linked from /machine/smram.  */
     cpu->machine_done.notify = x86_cpu_machine_done;
     qemu_add_machine_init_done_notifier(&cpu->machine_done);
+    return true;
 }
 
 #else /* CONFIG_USER_ONLY */
 
-static void tcg_cpu_realizefn(CPUState *cs, Error **errp)
+static bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
 {
+    return true;
 }
 
 #endif /* !CONFIG_USER_ONLY */
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 06/15] meson: add target_user_arch
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (4 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 05/15] accel-cpu: make cpu_realizefn return a bool Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:37   ` Alex Bennée
  2021-02-12 12:36 ` [RFC v18 07/15] i386: split off softmmu-only functionality in tcg-cpu Claudio Fontana
                   ` (10 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

the lack of target_user_arch makes it hard to fully leverage the
build system in order to separate user code from softmmu code.

Provide it, so that we can avoid the proliferation of #ifdef
in target code.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 meson.build                   | 5 +++++
 target/alpha/meson.build      | 3 +++
 target/arm/meson.build        | 2 ++
 target/cris/meson.build       | 3 +++
 target/hppa/meson.build       | 3 +++
 target/i386/meson.build       | 2 ++
 target/m68k/meson.build       | 3 +++
 target/microblaze/meson.build | 3 +++
 target/mips/meson.build       | 3 +++
 target/nios2/meson.build      | 3 +++
 target/openrisc/meson.build   | 3 +++
 target/ppc/meson.build        | 3 +++
 target/riscv/meson.build      | 3 +++
 target/s390x/meson.build      | 3 +++
 target/sh4/meson.build        | 3 +++
 target/sparc/meson.build      | 3 +++
 target/tilegx/meson.build     | 3 +++
 target/tricore/meson.build    | 3 +++
 target/xtensa/meson.build     | 3 +++
 19 files changed, 57 insertions(+)

diff --git a/meson.build b/meson.build
index a923f249d8..1a355b8c1e 100644
--- a/meson.build
+++ b/meson.build
@@ -1734,6 +1734,7 @@ modules = {}
 hw_arch = {}
 target_arch = {}
 target_softmmu_arch = {}
+target_user_arch = {}
 
 ###############
 # Trace files #
@@ -2131,6 +2132,10 @@ foreach target : target_dirs
     abi = config_target['TARGET_ABI_DIR']
     target_type='user'
     qemu_target_name = 'qemu-' + target_name
+    t = target_user_arch[arch].apply(config_target, strict: false)
+    arch_srcs += t.sources()
+    arch_deps += t.dependencies()
+
     if 'CONFIG_LINUX_USER' in config_target
       base_dir = 'linux-user'
       target_inc += include_directories('linux-user/host/' / config_host['ARCH'])
diff --git a/target/alpha/meson.build b/target/alpha/meson.build
index 1aec55abb4..1b0555d3ee 100644
--- a/target/alpha/meson.build
+++ b/target/alpha/meson.build
@@ -14,5 +14,8 @@ alpha_ss.add(files(
 alpha_softmmu_ss = ss.source_set()
 alpha_softmmu_ss.add(files('machine.c'))
 
+alpha_user_ss = ss.source_set()
+
 target_arch += {'alpha': alpha_ss}
 target_softmmu_arch += {'alpha': alpha_softmmu_ss}
+target_user_arch += {'alpha': alpha_user_ss}
diff --git a/target/arm/meson.build b/target/arm/meson.build
index 15b936c101..a96af5ee1b 100644
--- a/target/arm/meson.build
+++ b/target/arm/meson.build
@@ -53,6 +53,8 @@ arm_softmmu_ss.add(files(
   'monitor.c',
   'psci.c',
 ))
+arm_user_ss = ss.source_set()
 
 target_arch += {'arm': arm_ss}
 target_softmmu_arch += {'arm': arm_softmmu_ss}
+target_user_arch += {'arm': arm_user_ss}
diff --git a/target/cris/meson.build b/target/cris/meson.build
index 67c3793c85..7fd81e0348 100644
--- a/target/cris/meson.build
+++ b/target/cris/meson.build
@@ -10,5 +10,8 @@ cris_ss.add(files(
 cris_softmmu_ss = ss.source_set()
 cris_softmmu_ss.add(files('mmu.c', 'machine.c'))
 
+cris_user_ss = ss.source_set()
+
 target_arch += {'cris': cris_ss}
 target_softmmu_arch += {'cris': cris_softmmu_ss}
+target_user_arch += {'cris': cris_user_ss}
diff --git a/target/hppa/meson.build b/target/hppa/meson.build
index 8a7ff82efc..85ad314671 100644
--- a/target/hppa/meson.build
+++ b/target/hppa/meson.build
@@ -15,5 +15,8 @@ hppa_ss.add(files(
 hppa_softmmu_ss = ss.source_set()
 hppa_softmmu_ss.add(files('machine.c'))
 
+hppa_user_ss = ss.source_set()
+
 target_arch += {'hppa': hppa_ss}
 target_softmmu_arch += {'hppa': hppa_softmmu_ss}
+target_user_arch += {'hppa': hppa_user_ss}
diff --git a/target/i386/meson.build b/target/i386/meson.build
index fd24479590..cac26a4581 100644
--- a/target/i386/meson.build
+++ b/target/i386/meson.build
@@ -19,6 +19,7 @@ i386_softmmu_ss.add(files(
   'machine.c',
   'monitor.c',
 ))
+i386_user_ss = ss.source_set()
 
 subdir('kvm')
 subdir('hax')
@@ -28,3 +29,4 @@ subdir('tcg')
 
 target_arch += {'i386': i386_ss}
 target_softmmu_arch += {'i386': i386_softmmu_ss}
+target_user_arch += {'i386': i386_user_ss}
diff --git a/target/m68k/meson.build b/target/m68k/meson.build
index 05cd9fbd1e..b507682684 100644
--- a/target/m68k/meson.build
+++ b/target/m68k/meson.build
@@ -13,5 +13,8 @@ m68k_ss.add(files(
 m68k_softmmu_ss = ss.source_set()
 m68k_softmmu_ss.add(files('monitor.c'))
 
+m68k_user_ss = ss.source_set()
+
 target_arch += {'m68k': m68k_ss}
 target_softmmu_arch += {'m68k': m68k_softmmu_ss}
+target_user_arch += {'m68k': m68k_user_ss}
diff --git a/target/microblaze/meson.build b/target/microblaze/meson.build
index 05ee0ec163..52d8fcb0a3 100644
--- a/target/microblaze/meson.build
+++ b/target/microblaze/meson.build
@@ -16,5 +16,8 @@ microblaze_softmmu_ss.add(files(
   'machine.c',
 ))
 
+microblaze_user_ss = ss.source_set()
+
 target_arch += {'microblaze': microblaze_ss}
 target_softmmu_arch += {'microblaze': microblaze_softmmu_ss}
+target_user_arch += {'microblaze': microblaze_user_ss}
diff --git a/target/mips/meson.build b/target/mips/meson.build
index 9741545440..30843c1521 100644
--- a/target/mips/meson.build
+++ b/target/mips/meson.build
@@ -36,5 +36,8 @@ mips_softmmu_ss.add(when: 'CONFIG_TCG', if_true: files(
   'cp0_helper.c',
 ))
 
+mips_user_ss = ss.source_set()
+
 target_arch += {'mips': mips_ss}
 target_softmmu_arch += {'mips': mips_softmmu_ss}
+target_user_arch += {'mips': mips_user_ss}
diff --git a/target/nios2/meson.build b/target/nios2/meson.build
index e643917db1..00367056fa 100644
--- a/target/nios2/meson.build
+++ b/target/nios2/meson.build
@@ -11,5 +11,8 @@ nios2_ss.add(files(
 nios2_softmmu_ss = ss.source_set()
 nios2_softmmu_ss.add(files('monitor.c'))
 
+nios2_user_ss = ss.source_set()
+
 target_arch += {'nios2': nios2_ss}
 target_softmmu_arch += {'nios2': nios2_softmmu_ss}
+target_user_arch += {'nios2': nios2_user_ss}
diff --git a/target/openrisc/meson.build b/target/openrisc/meson.build
index 9774a58306..794a9e8161 100644
--- a/target/openrisc/meson.build
+++ b/target/openrisc/meson.build
@@ -19,5 +19,8 @@ openrisc_ss.add(files(
 openrisc_softmmu_ss = ss.source_set()
 openrisc_softmmu_ss.add(files('machine.c'))
 
+openrisc_user_ss = ss.source_set()
+
 target_arch += {'openrisc': openrisc_ss}
 target_softmmu_arch += {'openrisc': openrisc_softmmu_ss}
+target_user_arch += {'openrisc': openrisc_user_ss}
diff --git a/target/ppc/meson.build b/target/ppc/meson.build
index bbfef90e08..cdd69bf989 100644
--- a/target/ppc/meson.build
+++ b/target/ppc/meson.build
@@ -33,5 +33,8 @@ ppc_softmmu_ss.add(when: 'TARGET_PPC64', if_true: files(
   'mmu-radix64.c',
 ))
 
+ppc_user_ss = ss.source_set()
+
 target_arch += {'ppc': ppc_ss}
 target_softmmu_arch += {'ppc': ppc_softmmu_ss}
+target_user_arch += {'ppc': ppc_user_ss}
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index 14a5c62dac..a55851336b 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -31,5 +31,8 @@ riscv_softmmu_ss.add(files(
   'machine.c'
 ))
 
+riscv_user_ss = ss.source_set()
+
 target_arch += {'riscv': riscv_ss}
 target_softmmu_arch += {'riscv': riscv_softmmu_ss}
+target_user_arch += {'riscv': riscv_user_ss}
diff --git a/target/s390x/meson.build b/target/s390x/meson.build
index c42eadb7d2..1219f64112 100644
--- a/target/s390x/meson.build
+++ b/target/s390x/meson.build
@@ -58,5 +58,8 @@ if host_machine.cpu_family() == 's390x' and cc.has_link_argument('-Wl,--s390-pgs
                        if_true: declare_dependency(link_args: ['-Wl,--s390-pgste']))
 endif
 
+s390x_user_ss = ss.source_set()
+
 target_arch += {'s390x': s390x_ss}
 target_softmmu_arch += {'s390x': s390x_softmmu_ss}
+target_user_arch += {'s390x': s390x_user_ss}
diff --git a/target/sh4/meson.build b/target/sh4/meson.build
index 56a57576da..5a05729bc1 100644
--- a/target/sh4/meson.build
+++ b/target/sh4/meson.build
@@ -10,5 +10,8 @@ sh4_ss.add(files(
 sh4_softmmu_ss = ss.source_set()
 sh4_softmmu_ss.add(files('monitor.c'))
 
+sh4_user_ss = ss.source_set()
+
 target_arch += {'sh4': sh4_ss}
 target_softmmu_arch += {'sh4': sh4_softmmu_ss}
+target_user_arch += {'sh4': sh4_user_ss}
diff --git a/target/sparc/meson.build b/target/sparc/meson.build
index a3638b9503..cc77a77064 100644
--- a/target/sparc/meson.build
+++ b/target/sparc/meson.build
@@ -19,5 +19,8 @@ sparc_softmmu_ss.add(files(
   'monitor.c',
 ))
 
+sparc_user_ss = ss.source_set()
+
 target_arch += {'sparc': sparc_ss}
 target_softmmu_arch += {'sparc': sparc_softmmu_ss}
+target_user_arch += {'sparc': sparc_user_ss}
diff --git a/target/tilegx/meson.build b/target/tilegx/meson.build
index 678590439c..23dab8b9a1 100644
--- a/target/tilegx/meson.build
+++ b/target/tilegx/meson.build
@@ -9,5 +9,8 @@ tilegx_ss.add(zlib)
 
 tilegx_softmmu_ss = ss.source_set()
 
+tilegx_user_ss = ss.source_set()
+
 target_arch += {'tilegx': tilegx_ss}
 target_softmmu_arch += {'tilegx': tilegx_softmmu_ss}
+target_user_arch += {'tilegx': tilegx_user_ss}
diff --git a/target/tricore/meson.build b/target/tricore/meson.build
index 0ccc829517..7086ae1a22 100644
--- a/target/tricore/meson.build
+++ b/target/tricore/meson.build
@@ -11,5 +11,8 @@ tricore_ss.add(zlib)
 
 tricore_softmmu_ss = ss.source_set()
 
+tricore_user_ss = ss.source_set()
+
 target_arch += {'tricore': tricore_ss}
 target_softmmu_arch += {'tricore': tricore_softmmu_ss}
+target_user_arch += {'tricore': tricore_user_ss}
diff --git a/target/xtensa/meson.build b/target/xtensa/meson.build
index dd750a977e..949b2c8334 100644
--- a/target/xtensa/meson.build
+++ b/target/xtensa/meson.build
@@ -28,5 +28,8 @@ xtensa_softmmu_ss.add(files(
   'xtensa-semi.c',
 ))
 
+xtensa_user_ss = ss.source_set()
+
 target_arch += {'xtensa': xtensa_ss}
 target_softmmu_arch += {'xtensa': xtensa_softmmu_ss}
+target_user_arch += {'xtensa': xtensa_user_ss}
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 07/15] i386: split off softmmu-only functionality in tcg-cpu
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (5 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 06/15] meson: add target_user_arch Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 08/15] i386: split smm helper (softmmu) Claudio Fontana
                   ` (9 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/tcg/tcg-cpu.h           | 24 +++++++++
 target/i386/tcg/softmmu/tcg-cpu.c   | 83 +++++++++++++++++++++++++++++
 target/i386/tcg/tcg-cpu.c           | 75 ++------------------------
 target/i386/tcg/meson.build         |  3 ++
 target/i386/tcg/softmmu/meson.build |  3 ++
 target/i386/tcg/user/meson.build    |  2 +
 6 files changed, 119 insertions(+), 71 deletions(-)
 create mode 100644 target/i386/tcg/tcg-cpu.h
 create mode 100644 target/i386/tcg/softmmu/tcg-cpu.c
 create mode 100644 target/i386/tcg/softmmu/meson.build
 create mode 100644 target/i386/tcg/user/meson.build

diff --git a/target/i386/tcg/tcg-cpu.h b/target/i386/tcg/tcg-cpu.h
new file mode 100644
index 0000000000..36bd300af0
--- /dev/null
+++ b/target/i386/tcg/tcg-cpu.h
@@ -0,0 +1,24 @@
+/*
+ * i386 TCG cpu class initialization functions
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+#ifndef TCG_CPU_H
+#define TCG_CPU_H
+
+bool tcg_cpu_realizefn(CPUState *cs, Error **errp);
+
+#endif /* TCG_CPU_H */
diff --git a/target/i386/tcg/softmmu/tcg-cpu.c b/target/i386/tcg/softmmu/tcg-cpu.c
new file mode 100644
index 0000000000..2d45090c44
--- /dev/null
+++ b/target/i386/tcg/softmmu/tcg-cpu.c
@@ -0,0 +1,83 @@
+/*
+ * i386 TCG cpu class initialization functions specific to softmmu
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "tcg/helper-tcg.h"
+
+#include "sysemu/sysemu.h"
+#include "qemu/units.h"
+#include "exec/address-spaces.h"
+
+#include "tcg/tcg-cpu.h"
+
+static void tcg_cpu_machine_done(Notifier *n, void *unused)
+{
+    X86CPU *cpu = container_of(n, X86CPU, machine_done);
+    MemoryRegion *smram =
+        (MemoryRegion *) object_resolve_path("/machine/smram", NULL);
+
+    if (smram) {
+        cpu->smram = g_new(MemoryRegion, 1);
+        memory_region_init_alias(cpu->smram, OBJECT(cpu), "smram",
+                                 smram, 0, 4 * GiB);
+        memory_region_set_enabled(cpu->smram, true);
+        memory_region_add_subregion_overlap(cpu->cpu_as_root, 0,
+                                            cpu->smram, 1);
+    }
+}
+
+bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
+{
+    X86CPU *cpu = X86_CPU(cs);
+
+    /*
+     * The realize order is important, since x86_cpu_realize() checks if
+     * nothing else has been set by the user (or by accelerators) in
+     * cpu->ucode_rev and cpu->phys_bits, and the memory regions
+     * initialized here are needed for the vcpu initialization.
+     *
+     * realize order:
+     * tcg_cpu -> host_cpu -> x86_cpu
+     */
+    cpu->cpu_as_mem = g_new(MemoryRegion, 1);
+    cpu->cpu_as_root = g_new(MemoryRegion, 1);
+
+    /* Outer container... */
+    memory_region_init(cpu->cpu_as_root, OBJECT(cpu), "memory", ~0ull);
+    memory_region_set_enabled(cpu->cpu_as_root, true);
+
+    /*
+     * ... with two regions inside: normal system memory with low
+     * priority, and...
+     */
+    memory_region_init_alias(cpu->cpu_as_mem, OBJECT(cpu), "memory",
+                             get_system_memory(), 0, ~0ull);
+    memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);
+    memory_region_set_enabled(cpu->cpu_as_mem, true);
+
+    cs->num_ases = 2;
+    cpu_address_space_init(cs, 0, "cpu-memory", cs->memory);
+    cpu_address_space_init(cs, 1, "cpu-smm", cpu->cpu_as_root);
+
+    /* ... SMRAM with higher priority, linked from /machine/smram.  */
+    cpu->machine_done.notify = tcg_cpu_machine_done;
+    qemu_add_machine_init_done_notifier(&cpu->machine_done);
+    return true;
+}
diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c
index 23e1f5f0c3..e311f52855 100644
--- a/target/i386/tcg/tcg-cpu.c
+++ b/target/i386/tcg/tcg-cpu.c
@@ -23,11 +23,7 @@
 #include "qemu/accel.h"
 #include "hw/core/accel-cpu.h"
 
-#ifndef CONFIG_USER_ONLY
-#include "sysemu/sysemu.h"
-#include "qemu/units.h"
-#include "exec/address-spaces.h"
-#endif
+#include "tcg-cpu.h"
 
 /* Frob eflags into and out of the CPU temporary format.  */
 
@@ -78,72 +74,6 @@ static void tcg_cpu_class_init(CPUClass *cc)
     cc->tcg_ops = &x86_tcg_ops;
 }
 
-#ifndef CONFIG_USER_ONLY
-
-static void x86_cpu_machine_done(Notifier *n, void *unused)
-{
-    X86CPU *cpu = container_of(n, X86CPU, machine_done);
-    MemoryRegion *smram =
-        (MemoryRegion *) object_resolve_path("/machine/smram", NULL);
-
-    if (smram) {
-        cpu->smram = g_new(MemoryRegion, 1);
-        memory_region_init_alias(cpu->smram, OBJECT(cpu), "smram",
-                                 smram, 0, 4 * GiB);
-        memory_region_set_enabled(cpu->smram, true);
-        memory_region_add_subregion_overlap(cpu->cpu_as_root, 0,
-                                            cpu->smram, 1);
-    }
-}
-
-static bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
-{
-    X86CPU *cpu = X86_CPU(cs);
-
-    /*
-     * The realize order is important, since x86_cpu_realize() checks if
-     * nothing else has been set by the user (or by accelerators) in
-     * cpu->ucode_rev and cpu->phys_bits, and the memory regions
-     * initialized here are needed for the vcpu initialization.
-     *
-     * realize order:
-     * tcg_cpu -> host_cpu -> x86_cpu
-     */
-    cpu->cpu_as_mem = g_new(MemoryRegion, 1);
-    cpu->cpu_as_root = g_new(MemoryRegion, 1);
-
-    /* Outer container... */
-    memory_region_init(cpu->cpu_as_root, OBJECT(cpu), "memory", ~0ull);
-    memory_region_set_enabled(cpu->cpu_as_root, true);
-
-    /*
-     * ... with two regions inside: normal system memory with low
-     * priority, and...
-     */
-    memory_region_init_alias(cpu->cpu_as_mem, OBJECT(cpu), "memory",
-                             get_system_memory(), 0, ~0ull);
-    memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);
-    memory_region_set_enabled(cpu->cpu_as_mem, true);
-
-    cs->num_ases = 2;
-    cpu_address_space_init(cs, 0, "cpu-memory", cs->memory);
-    cpu_address_space_init(cs, 1, "cpu-smm", cpu->cpu_as_root);
-
-    /* ... SMRAM with higher priority, linked from /machine/smram.  */
-    cpu->machine_done.notify = x86_cpu_machine_done;
-    qemu_add_machine_init_done_notifier(&cpu->machine_done);
-    return true;
-}
-
-#else /* CONFIG_USER_ONLY */
-
-static bool tcg_cpu_realizefn(CPUState *cs, Error **errp)
-{
-    return true;
-}
-
-#endif /* !CONFIG_USER_ONLY */
-
 /*
  * TCG-specific defaults that override all CPU models when using TCG
  */
@@ -163,7 +93,10 @@ static void tcg_cpu_accel_class_init(ObjectClass *oc, void *data)
 {
     AccelCPUClass *acc = ACCEL_CPU_CLASS(oc);
 
+#ifndef CONFIG_USER_ONLY
     acc->cpu_realizefn = tcg_cpu_realizefn;
+#endif /* CONFIG_USER_ONLY */
+
     acc->cpu_class_init = tcg_cpu_class_init;
     acc->cpu_instance_init = tcg_cpu_instance_init;
 }
diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
index 6a1a73cdbf..68fa0c3187 100644
--- a/target/i386/tcg/meson.build
+++ b/target/i386/tcg/meson.build
@@ -12,3 +12,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
   'svm_helper.c',
   'tcg-cpu.c',
   'translate.c'), if_false: files('tcg-stub.c'))
+
+subdir('softmmu')
+subdir('user')
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
new file mode 100644
index 0000000000..4ab30cc32e
--- /dev/null
+++ b/target/i386/tcg/softmmu/meson.build
@@ -0,0 +1,3 @@
+i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
+  'tcg-cpu.c',
+))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
new file mode 100644
index 0000000000..7aecc53155
--- /dev/null
+++ b/target/i386/tcg/user/meson.build
@@ -0,0 +1,2 @@
+i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
+))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (6 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 07/15] i386: split off softmmu-only functionality in tcg-cpu Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:51   ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 09/15] i386: split tcg excp_helper into softmmu and user parts Claudio Fontana
                   ` (8 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

smm is only really useful for softmmu, split in two modules
around the CONFIG_USER_ONLY, in order to remove the ifdef
and use the build system instead.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/helper.h                       |  4 ++++
 target/i386/tcg/seg_helper.c               |  2 ++
 target/i386/tcg/{ => softmmu}/smm_helper.c | 19 ++-----------------
 target/i386/tcg/translate.c                |  2 ++
 target/i386/tcg/meson.build                |  1 -
 target/i386/tcg/softmmu/meson.build        |  1 +
 6 files changed, 11 insertions(+), 18 deletions(-)
 rename target/i386/tcg/{ => softmmu}/smm_helper.c (98%)

diff --git a/target/i386/helper.h b/target/i386/helper.h
index c2ae2f7e61..8ffda4cdc6 100644
--- a/target/i386/helper.h
+++ b/target/i386/helper.h
@@ -70,7 +70,11 @@ DEF_HELPER_1(clac, void, env)
 DEF_HELPER_1(stac, void, env)
 DEF_HELPER_3(boundw, void, env, tl, int)
 DEF_HELPER_3(boundl, void, env, tl, int)
+
+#ifndef CONFIG_USER_ONLY
 DEF_HELPER_1(rsm, void, env)
+#endif /* !CONFIG_USER_ONLY */
+
 DEF_HELPER_2(into, void, env, int)
 DEF_HELPER_2(cmpxchg8b_unlocked, void, env, tl)
 DEF_HELPER_2(cmpxchg8b, void, env, tl)
diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
index 180d47f0e9..f0cb1bffe7 100644
--- a/target/i386/tcg/seg_helper.c
+++ b/target/i386/tcg/seg_helper.c
@@ -1351,7 +1351,9 @@ bool x86_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
     case CPU_INTERRUPT_SMI:
         cpu_svm_check_intercept_param(env, SVM_EXIT_SMI, 0, 0);
         cs->interrupt_request &= ~CPU_INTERRUPT_SMI;
+#ifndef CONFIG_USER_ONLY
         do_smm_enter(cpu);
+#endif
         break;
     case CPU_INTERRUPT_NMI:
         cpu_svm_check_intercept_param(env, SVM_EXIT_NMI, 0, 0);
diff --git a/target/i386/tcg/smm_helper.c b/target/i386/tcg/softmmu/smm_helper.c
similarity index 98%
rename from target/i386/tcg/smm_helper.c
rename to target/i386/tcg/softmmu/smm_helper.c
index 62d027abd3..ee53b26629 100644
--- a/target/i386/tcg/smm_helper.c
+++ b/target/i386/tcg/softmmu/smm_helper.c
@@ -1,5 +1,5 @@
 /*
- *  x86 SMM helpers
+ *  x86 SMM helpers (softmmu-only)
  *
  *  Copyright (c) 2003 Fabrice Bellard
  *
@@ -18,27 +18,14 @@
  */
 
 #include "qemu/osdep.h"
-#include "qemu/main-loop.h"
 #include "cpu.h"
 #include "exec/helper-proto.h"
 #include "exec/log.h"
-#include "helper-tcg.h"
+#include "tcg/helper-tcg.h"
 
 
 /* SMM support */
 
-#if defined(CONFIG_USER_ONLY)
-
-void do_smm_enter(X86CPU *cpu)
-{
-}
-
-void helper_rsm(CPUX86State *env)
-{
-}
-
-#else
-
 #ifdef TARGET_X86_64
 #define SMM_REVISION_ID 0x00020064
 #else
@@ -330,5 +317,3 @@ void helper_rsm(CPUX86State *env)
     qemu_log_mask(CPU_LOG_INT, "SMM: after RSM\n");
     log_cpu_state_mask(CPU_LOG_INT, CPU(cpu), CPU_DUMP_CCOP);
 }
-
-#endif /* !CONFIG_USER_ONLY */
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index af1faf9342..5075ac4830 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -8321,7 +8321,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
             goto illegal_op;
         gen_update_cc_op(s);
         gen_jmp_im(s, s->pc - s->cs_base);
+#ifndef CONFIG_USER_ONLY
         gen_helper_rsm(cpu_env);
+#endif /* CONFIG_USER_ONLY */
         gen_eob(s);
         break;
     case 0x1b8: /* SSE4.2 popcnt */
diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
index 68fa0c3187..ec5daa1edc 100644
--- a/target/i386/tcg/meson.build
+++ b/target/i386/tcg/meson.build
@@ -8,7 +8,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
   'misc_helper.c',
   'mpx_helper.c',
   'seg_helper.c',
-  'smm_helper.c',
   'svm_helper.c',
   'tcg-cpu.c',
   'translate.c'), if_false: files('tcg-stub.c'))
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index 4ab30cc32e..35ba16dc3d 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -1,3 +1,4 @@
 i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'tcg-cpu.c',
+  'smm_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 09/15] i386: split tcg excp_helper into softmmu and user parts
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (7 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 08/15] i386: split smm helper (softmmu) Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 10/15] i386: split tcg btp_helper " Claudio Fontana
                   ` (7 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/tcg/excp_helper.c         | 572 --------------------------
 target/i386/tcg/softmmu/excp_helper.c | 562 +++++++++++++++++++++++++
 target/i386/tcg/user/excp_helper.c    |  39 ++
 target/i386/tcg/softmmu/meson.build   |   1 +
 target/i386/tcg/user/meson.build      |   1 +
 5 files changed, 603 insertions(+), 572 deletions(-)
 create mode 100644 target/i386/tcg/softmmu/excp_helper.c
 create mode 100644 target/i386/tcg/user/excp_helper.c

diff --git a/target/i386/tcg/excp_helper.c b/target/i386/tcg/excp_helper.c
index b7d6259e4a..0183f3932e 100644
--- a/target/i386/tcg/excp_helper.c
+++ b/target/i386/tcg/excp_helper.c
@@ -137,575 +137,3 @@ void raise_exception_ra(CPUX86State *env, int exception_index, uintptr_t retaddr
 {
     raise_interrupt2(env, exception_index, 0, 0, 0, retaddr);
 }
-
-#if !defined(CONFIG_USER_ONLY)
-static hwaddr get_hphys(CPUState *cs, hwaddr gphys, MMUAccessType access_type,
-                        int *prot)
-{
-    CPUX86State *env = &X86_CPU(cs)->env;
-    uint64_t rsvd_mask = PG_HI_RSVD_MASK;
-    uint64_t ptep, pte;
-    uint64_t exit_info_1 = 0;
-    target_ulong pde_addr, pte_addr;
-    uint32_t page_offset;
-    int page_size;
-
-    if (likely(!(env->hflags2 & HF2_NPT_MASK))) {
-        return gphys;
-    }
-
-    if (!(env->nested_pg_mode & SVM_NPT_NXE)) {
-        rsvd_mask |= PG_NX_MASK;
-    }
-
-    if (env->nested_pg_mode & SVM_NPT_PAE) {
-        uint64_t pde, pdpe;
-        target_ulong pdpe_addr;
-
-#ifdef TARGET_X86_64
-        if (env->nested_pg_mode & SVM_NPT_LMA) {
-            uint64_t pml5e;
-            uint64_t pml4e_addr, pml4e;
-
-            pml5e = env->nested_cr3;
-            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
-
-            pml4e_addr = (pml5e & PG_ADDRESS_MASK) +
-                    (((gphys >> 39) & 0x1ff) << 3);
-            pml4e = x86_ldq_phys(cs, pml4e_addr);
-            if (!(pml4e & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
-                goto do_fault_rsvd;
-            }
-            if (!(pml4e & PG_ACCESSED_MASK)) {
-                pml4e |= PG_ACCESSED_MASK;
-                x86_stl_phys_notdirty(cs, pml4e_addr, pml4e);
-            }
-            ptep &= pml4e ^ PG_NX_MASK;
-            pdpe_addr = (pml4e & PG_ADDRESS_MASK) +
-                    (((gphys >> 30) & 0x1ff) << 3);
-            pdpe = x86_ldq_phys(cs, pdpe_addr);
-            if (!(pdpe & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            if (pdpe & rsvd_mask) {
-                goto do_fault_rsvd;
-            }
-            ptep &= pdpe ^ PG_NX_MASK;
-            if (!(pdpe & PG_ACCESSED_MASK)) {
-                pdpe |= PG_ACCESSED_MASK;
-                x86_stl_phys_notdirty(cs, pdpe_addr, pdpe);
-            }
-            if (pdpe & PG_PSE_MASK) {
-                /* 1 GB page */
-                page_size = 1024 * 1024 * 1024;
-                pte_addr = pdpe_addr;
-                pte = pdpe;
-                goto do_check_protect;
-            }
-        } else
-#endif
-        {
-            pdpe_addr = (env->nested_cr3 & ~0x1f) + ((gphys >> 27) & 0x18);
-            pdpe = x86_ldq_phys(cs, pdpe_addr);
-            if (!(pdpe & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            rsvd_mask |= PG_HI_USER_MASK;
-            if (pdpe & (rsvd_mask | PG_NX_MASK)) {
-                goto do_fault_rsvd;
-            }
-            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
-        }
-
-        pde_addr = (pdpe & PG_ADDRESS_MASK) + (((gphys >> 21) & 0x1ff) << 3);
-        pde = x86_ldq_phys(cs, pde_addr);
-        if (!(pde & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        if (pde & rsvd_mask) {
-            goto do_fault_rsvd;
-        }
-        ptep &= pde ^ PG_NX_MASK;
-        if (pde & PG_PSE_MASK) {
-            /* 2 MB page */
-            page_size = 2048 * 1024;
-            pte_addr = pde_addr;
-            pte = pde;
-            goto do_check_protect;
-        }
-        /* 4 KB page */
-        if (!(pde & PG_ACCESSED_MASK)) {
-            pde |= PG_ACCESSED_MASK;
-            x86_stl_phys_notdirty(cs, pde_addr, pde);
-        }
-        pte_addr = (pde & PG_ADDRESS_MASK) + (((gphys >> 12) & 0x1ff) << 3);
-        pte = x86_ldq_phys(cs, pte_addr);
-        if (!(pte & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        if (pte & rsvd_mask) {
-            goto do_fault_rsvd;
-        }
-        /* combine pde and pte nx, user and rw protections */
-        ptep &= pte ^ PG_NX_MASK;
-        page_size = 4096;
-    } else {
-        uint32_t pde;
-
-        /* page directory entry */
-        pde_addr = (env->nested_cr3 & ~0xfff) + ((gphys >> 20) & 0xffc);
-        pde = x86_ldl_phys(cs, pde_addr);
-        if (!(pde & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        ptep = pde | PG_NX_MASK;
-
-        /* if host cr4 PSE bit is set, then we use a 4MB page */
-        if ((pde & PG_PSE_MASK) && (env->nested_pg_mode & SVM_NPT_PSE)) {
-            page_size = 4096 * 1024;
-            pte_addr = pde_addr;
-
-            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.
-             * Leave bits 20-13 in place for setting accessed/dirty bits below.
-             */
-            pte = pde | ((pde & 0x1fe000LL) << (32 - 13));
-            rsvd_mask = 0x200000;
-            goto do_check_protect_pse36;
-        }
-
-        if (!(pde & PG_ACCESSED_MASK)) {
-            pde |= PG_ACCESSED_MASK;
-            x86_stl_phys_notdirty(cs, pde_addr, pde);
-        }
-
-        /* page directory entry */
-        pte_addr = (pde & ~0xfff) + ((gphys >> 10) & 0xffc);
-        pte = x86_ldl_phys(cs, pte_addr);
-        if (!(pte & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        /* combine pde and pte user and rw protections */
-        ptep &= pte | PG_NX_MASK;
-        page_size = 4096;
-        rsvd_mask = 0;
-    }
-
- do_check_protect:
-    rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK;
- do_check_protect_pse36:
-    if (pte & rsvd_mask) {
-        goto do_fault_rsvd;
-    }
-    ptep ^= PG_NX_MASK;
-
-    if (!(ptep & PG_USER_MASK)) {
-        goto do_fault_protect;
-    }
-    if (ptep & PG_NX_MASK) {
-        if (access_type == MMU_INST_FETCH) {
-            goto do_fault_protect;
-        }
-        *prot &= ~PAGE_EXEC;
-    }
-    if (!(ptep & PG_RW_MASK)) {
-        if (access_type == MMU_DATA_STORE) {
-            goto do_fault_protect;
-        }
-        *prot &= ~PAGE_WRITE;
-    }
-
-    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
-    page_offset = gphys & (page_size - 1);
-    return pte + page_offset;
-
- do_fault_rsvd:
-    exit_info_1 |= SVM_NPTEXIT_RSVD;
- do_fault_protect:
-    exit_info_1 |= SVM_NPTEXIT_P;
- do_fault:
-    x86_stq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2),
-                 gphys);
-    exit_info_1 |= SVM_NPTEXIT_US;
-    if (access_type == MMU_DATA_STORE) {
-        exit_info_1 |= SVM_NPTEXIT_RW;
-    } else if (access_type == MMU_INST_FETCH) {
-        exit_info_1 |= SVM_NPTEXIT_ID;
-    }
-    if (prot) {
-        exit_info_1 |= SVM_NPTEXIT_GPA;
-    } else { /* page table access */
-        exit_info_1 |= SVM_NPTEXIT_GPT;
-    }
-    cpu_vmexit(env, SVM_EXIT_NPF, exit_info_1, env->retaddr);
-}
-
-/* return value:
- * -1 = cannot handle fault
- * 0  = nothing more to do
- * 1  = generate PF fault
- */
-static int handle_mmu_fault(CPUState *cs, vaddr addr, int size,
-                            int is_write1, int mmu_idx)
-{
-    X86CPU *cpu = X86_CPU(cs);
-    CPUX86State *env = &cpu->env;
-    uint64_t ptep, pte;
-    int32_t a20_mask;
-    target_ulong pde_addr, pte_addr;
-    int error_code = 0;
-    int is_dirty, prot, page_size, is_write, is_user;
-    hwaddr paddr;
-    uint64_t rsvd_mask = PG_HI_RSVD_MASK;
-    uint32_t page_offset;
-    target_ulong vaddr;
-    uint32_t pkr;
-
-    is_user = mmu_idx == MMU_USER_IDX;
-#if defined(DEBUG_MMU)
-    printf("MMU fault: addr=%" VADDR_PRIx " w=%d u=%d eip=" TARGET_FMT_lx "\n",
-           addr, is_write1, is_user, env->eip);
-#endif
-    is_write = is_write1 & 1;
-
-    a20_mask = x86_get_a20_mask(env);
-    if (!(env->cr[0] & CR0_PG_MASK)) {
-        pte = addr;
-#ifdef TARGET_X86_64
-        if (!(env->hflags & HF_LMA_MASK)) {
-            /* Without long mode we can only address 32bits in real mode */
-            pte = (uint32_t)pte;
-        }
-#endif
-        prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
-        page_size = 4096;
-        goto do_mapping;
-    }
-
-    if (!(env->efer & MSR_EFER_NXE)) {
-        rsvd_mask |= PG_NX_MASK;
-    }
-
-    if (env->cr[4] & CR4_PAE_MASK) {
-        uint64_t pde, pdpe;
-        target_ulong pdpe_addr;
-
-#ifdef TARGET_X86_64
-        if (env->hflags & HF_LMA_MASK) {
-            bool la57 = env->cr[4] & CR4_LA57_MASK;
-            uint64_t pml5e_addr, pml5e;
-            uint64_t pml4e_addr, pml4e;
-            int32_t sext;
-
-            /* test virtual address sign extension */
-            sext = la57 ? (int64_t)addr >> 56 : (int64_t)addr >> 47;
-            if (sext != 0 && sext != -1) {
-                env->error_code = 0;
-                cs->exception_index = EXCP0D_GPF;
-                return 1;
-            }
-
-            if (la57) {
-                pml5e_addr = ((env->cr[3] & ~0xfff) +
-                        (((addr >> 48) & 0x1ff) << 3)) & a20_mask;
-                pml5e_addr = get_hphys(cs, pml5e_addr, MMU_DATA_STORE, NULL);
-                pml5e = x86_ldq_phys(cs, pml5e_addr);
-                if (!(pml5e & PG_PRESENT_MASK)) {
-                    goto do_fault;
-                }
-                if (pml5e & (rsvd_mask | PG_PSE_MASK)) {
-                    goto do_fault_rsvd;
-                }
-                if (!(pml5e & PG_ACCESSED_MASK)) {
-                    pml5e |= PG_ACCESSED_MASK;
-                    x86_stl_phys_notdirty(cs, pml5e_addr, pml5e);
-                }
-                ptep = pml5e ^ PG_NX_MASK;
-            } else {
-                pml5e = env->cr[3];
-                ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
-            }
-
-            pml4e_addr = ((pml5e & PG_ADDRESS_MASK) +
-                    (((addr >> 39) & 0x1ff) << 3)) & a20_mask;
-            pml4e_addr = get_hphys(cs, pml4e_addr, MMU_DATA_STORE, false);
-            pml4e = x86_ldq_phys(cs, pml4e_addr);
-            if (!(pml4e & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
-                goto do_fault_rsvd;
-            }
-            if (!(pml4e & PG_ACCESSED_MASK)) {
-                pml4e |= PG_ACCESSED_MASK;
-                x86_stl_phys_notdirty(cs, pml4e_addr, pml4e);
-            }
-            ptep &= pml4e ^ PG_NX_MASK;
-            pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) &
-                a20_mask;
-            pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, NULL);
-            pdpe = x86_ldq_phys(cs, pdpe_addr);
-            if (!(pdpe & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            if (pdpe & rsvd_mask) {
-                goto do_fault_rsvd;
-            }
-            ptep &= pdpe ^ PG_NX_MASK;
-            if (!(pdpe & PG_ACCESSED_MASK)) {
-                pdpe |= PG_ACCESSED_MASK;
-                x86_stl_phys_notdirty(cs, pdpe_addr, pdpe);
-            }
-            if (pdpe & PG_PSE_MASK) {
-                /* 1 GB page */
-                page_size = 1024 * 1024 * 1024;
-                pte_addr = pdpe_addr;
-                pte = pdpe;
-                goto do_check_protect;
-            }
-        } else
-#endif
-        {
-            /* XXX: load them when cr3 is loaded ? */
-            pdpe_addr = ((env->cr[3] & ~0x1f) + ((addr >> 27) & 0x18)) &
-                a20_mask;
-            pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, false);
-            pdpe = x86_ldq_phys(cs, pdpe_addr);
-            if (!(pdpe & PG_PRESENT_MASK)) {
-                goto do_fault;
-            }
-            rsvd_mask |= PG_HI_USER_MASK;
-            if (pdpe & (rsvd_mask | PG_NX_MASK)) {
-                goto do_fault_rsvd;
-            }
-            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
-        }
-
-        pde_addr = ((pdpe & PG_ADDRESS_MASK) + (((addr >> 21) & 0x1ff) << 3)) &
-            a20_mask;
-        pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL);
-        pde = x86_ldq_phys(cs, pde_addr);
-        if (!(pde & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        if (pde & rsvd_mask) {
-            goto do_fault_rsvd;
-        }
-        ptep &= pde ^ PG_NX_MASK;
-        if (pde & PG_PSE_MASK) {
-            /* 2 MB page */
-            page_size = 2048 * 1024;
-            pte_addr = pde_addr;
-            pte = pde;
-            goto do_check_protect;
-        }
-        /* 4 KB page */
-        if (!(pde & PG_ACCESSED_MASK)) {
-            pde |= PG_ACCESSED_MASK;
-            x86_stl_phys_notdirty(cs, pde_addr, pde);
-        }
-        pte_addr = ((pde & PG_ADDRESS_MASK) + (((addr >> 12) & 0x1ff) << 3)) &
-            a20_mask;
-        pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL);
-        pte = x86_ldq_phys(cs, pte_addr);
-        if (!(pte & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        if (pte & rsvd_mask) {
-            goto do_fault_rsvd;
-        }
-        /* combine pde and pte nx, user and rw protections */
-        ptep &= pte ^ PG_NX_MASK;
-        page_size = 4096;
-    } else {
-        uint32_t pde;
-
-        /* page directory entry */
-        pde_addr = ((env->cr[3] & ~0xfff) + ((addr >> 20) & 0xffc)) &
-            a20_mask;
-        pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL);
-        pde = x86_ldl_phys(cs, pde_addr);
-        if (!(pde & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        ptep = pde | PG_NX_MASK;
-
-        /* if PSE bit is set, then we use a 4MB page */
-        if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) {
-            page_size = 4096 * 1024;
-            pte_addr = pde_addr;
-
-            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.
-             * Leave bits 20-13 in place for setting accessed/dirty bits below.
-             */
-            pte = pde | ((pde & 0x1fe000LL) << (32 - 13));
-            rsvd_mask = 0x200000;
-            goto do_check_protect_pse36;
-        }
-
-        if (!(pde & PG_ACCESSED_MASK)) {
-            pde |= PG_ACCESSED_MASK;
-            x86_stl_phys_notdirty(cs, pde_addr, pde);
-        }
-
-        /* page directory entry */
-        pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) &
-            a20_mask;
-        pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL);
-        pte = x86_ldl_phys(cs, pte_addr);
-        if (!(pte & PG_PRESENT_MASK)) {
-            goto do_fault;
-        }
-        /* combine pde and pte user and rw protections */
-        ptep &= pte | PG_NX_MASK;
-        page_size = 4096;
-        rsvd_mask = 0;
-    }
-
-do_check_protect:
-    rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK;
-do_check_protect_pse36:
-    if (pte & rsvd_mask) {
-        goto do_fault_rsvd;
-    }
-    ptep ^= PG_NX_MASK;
-
-    /* can the page can be put in the TLB?  prot will tell us */
-    if (is_user && !(ptep & PG_USER_MASK)) {
-        goto do_fault_protect;
-    }
-
-    prot = 0;
-    if (mmu_idx != MMU_KSMAP_IDX || !(ptep & PG_USER_MASK)) {
-        prot |= PAGE_READ;
-        if ((ptep & PG_RW_MASK) || (!is_user && !(env->cr[0] & CR0_WP_MASK))) {
-            prot |= PAGE_WRITE;
-        }
-    }
-    if (!(ptep & PG_NX_MASK) &&
-        (mmu_idx == MMU_USER_IDX ||
-         !((env->cr[4] & CR4_SMEP_MASK) && (ptep & PG_USER_MASK)))) {
-        prot |= PAGE_EXEC;
-    }
-
-    if (!(env->hflags & HF_LMA_MASK)) {
-        pkr = 0;
-    } else if (ptep & PG_USER_MASK) {
-        pkr = env->cr[4] & CR4_PKE_MASK ? env->pkru : 0;
-    } else {
-        pkr = env->cr[4] & CR4_PKS_MASK ? env->pkrs : 0;
-    }
-    if (pkr) {
-        uint32_t pk = (pte & PG_PKRU_MASK) >> PG_PKRU_BIT;
-        uint32_t pkr_ad = (pkr >> pk * 2) & 1;
-        uint32_t pkr_wd = (pkr >> pk * 2) & 2;
-        uint32_t pkr_prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
-
-        if (pkr_ad) {
-            pkr_prot &= ~(PAGE_READ | PAGE_WRITE);
-        } else if (pkr_wd && (is_user || env->cr[0] & CR0_WP_MASK)) {
-            pkr_prot &= ~PAGE_WRITE;
-        }
-
-        prot &= pkr_prot;
-        if ((pkr_prot & (1 << is_write1)) == 0) {
-            assert(is_write1 != 2);
-            error_code |= PG_ERROR_PK_MASK;
-            goto do_fault_protect;
-        }
-    }
-
-    if ((prot & (1 << is_write1)) == 0) {
-        goto do_fault_protect;
-    }
-
-    /* yes, it can! */
-    is_dirty = is_write && !(pte & PG_DIRTY_MASK);
-    if (!(pte & PG_ACCESSED_MASK) || is_dirty) {
-        pte |= PG_ACCESSED_MASK;
-        if (is_dirty) {
-            pte |= PG_DIRTY_MASK;
-        }
-        x86_stl_phys_notdirty(cs, pte_addr, pte);
-    }
-
-    if (!(pte & PG_DIRTY_MASK)) {
-        /* only set write access if already dirty... otherwise wait
-           for dirty access */
-        assert(!is_write);
-        prot &= ~PAGE_WRITE;
-    }
-
- do_mapping:
-    pte = pte & a20_mask;
-
-    /* align to page_size */
-    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
-    page_offset = addr & (page_size - 1);
-    paddr = get_hphys(cs, pte + page_offset, is_write1, &prot);
-
-    /* Even if 4MB pages, we map only one 4KB page in the cache to
-       avoid filling it too fast */
-    vaddr = addr & TARGET_PAGE_MASK;
-    paddr &= TARGET_PAGE_MASK;
-
-    assert(prot & (1 << is_write1));
-    tlb_set_page_with_attrs(cs, vaddr, paddr, cpu_get_mem_attrs(env),
-                            prot, mmu_idx, page_size);
-    return 0;
- do_fault_rsvd:
-    error_code |= PG_ERROR_RSVD_MASK;
- do_fault_protect:
-    error_code |= PG_ERROR_P_MASK;
- do_fault:
-    error_code |= (is_write << PG_ERROR_W_BIT);
-    if (is_user)
-        error_code |= PG_ERROR_U_MASK;
-    if (is_write1 == 2 &&
-        (((env->efer & MSR_EFER_NXE) &&
-          (env->cr[4] & CR4_PAE_MASK)) ||
-         (env->cr[4] & CR4_SMEP_MASK)))
-        error_code |= PG_ERROR_I_D_MASK;
-    if (env->intercept_exceptions & (1 << EXCP0E_PAGE)) {
-        /* cr2 is not modified in case of exceptions */
-        x86_stq_phys(cs,
-                 env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2),
-                 addr);
-    } else {
-        env->cr[2] = addr;
-    }
-    env->error_code = error_code;
-    cs->exception_index = EXCP0E_PAGE;
-    return 1;
-}
-#endif
-
-bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size,
-                      MMUAccessType access_type, int mmu_idx,
-                      bool probe, uintptr_t retaddr)
-{
-    X86CPU *cpu = X86_CPU(cs);
-    CPUX86State *env = &cpu->env;
-
-#ifdef CONFIG_USER_ONLY
-    /* user mode only emulation */
-    env->cr[2] = addr;
-    env->error_code = (access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT;
-    env->error_code |= PG_ERROR_U_MASK;
-    cs->exception_index = EXCP0E_PAGE;
-    env->exception_is_int = 0;
-    env->exception_next_eip = -1;
-    cpu_loop_exit_restore(cs, retaddr);
-#else
-    env->retaddr = retaddr;
-    if (handle_mmu_fault(cs, addr, size, access_type, mmu_idx)) {
-        /* FIXME: On error in get_hphys we have already jumped out.  */
-        g_assert(!probe);
-        raise_exception_err_ra(env, cs->exception_index,
-                               env->error_code, retaddr);
-    }
-    return true;
-#endif
-}
diff --git a/target/i386/tcg/softmmu/excp_helper.c b/target/i386/tcg/softmmu/excp_helper.c
new file mode 100644
index 0000000000..fac9ec74d5
--- /dev/null
+++ b/target/i386/tcg/softmmu/excp_helper.c
@@ -0,0 +1,562 @@
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "tcg/helper-tcg.h"
+
+static hwaddr get_hphys(CPUState *cs, hwaddr gphys, MMUAccessType access_type,
+                        int *prot)
+{
+    CPUX86State *env = &X86_CPU(cs)->env;
+    uint64_t rsvd_mask = PG_HI_RSVD_MASK;
+    uint64_t ptep, pte;
+    uint64_t exit_info_1 = 0;
+    target_ulong pde_addr, pte_addr;
+    uint32_t page_offset;
+    int page_size;
+
+    if (likely(!(env->hflags2 & HF2_NPT_MASK))) {
+        return gphys;
+    }
+
+    if (!(env->nested_pg_mode & SVM_NPT_NXE)) {
+        rsvd_mask |= PG_NX_MASK;
+    }
+
+    if (env->nested_pg_mode & SVM_NPT_PAE) {
+        uint64_t pde, pdpe;
+        target_ulong pdpe_addr;
+
+#ifdef TARGET_X86_64
+        if (env->nested_pg_mode & SVM_NPT_LMA) {
+            uint64_t pml5e;
+            uint64_t pml4e_addr, pml4e;
+
+            pml5e = env->nested_cr3;
+            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
+
+            pml4e_addr = (pml5e & PG_ADDRESS_MASK) +
+                    (((gphys >> 39) & 0x1ff) << 3);
+            pml4e = x86_ldq_phys(cs, pml4e_addr);
+            if (!(pml4e & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
+                goto do_fault_rsvd;
+            }
+            if (!(pml4e & PG_ACCESSED_MASK)) {
+                pml4e |= PG_ACCESSED_MASK;
+                x86_stl_phys_notdirty(cs, pml4e_addr, pml4e);
+            }
+            ptep &= pml4e ^ PG_NX_MASK;
+            pdpe_addr = (pml4e & PG_ADDRESS_MASK) +
+                    (((gphys >> 30) & 0x1ff) << 3);
+            pdpe = x86_ldq_phys(cs, pdpe_addr);
+            if (!(pdpe & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            if (pdpe & rsvd_mask) {
+                goto do_fault_rsvd;
+            }
+            ptep &= pdpe ^ PG_NX_MASK;
+            if (!(pdpe & PG_ACCESSED_MASK)) {
+                pdpe |= PG_ACCESSED_MASK;
+                x86_stl_phys_notdirty(cs, pdpe_addr, pdpe);
+            }
+            if (pdpe & PG_PSE_MASK) {
+                /* 1 GB page */
+                page_size = 1024 * 1024 * 1024;
+                pte_addr = pdpe_addr;
+                pte = pdpe;
+                goto do_check_protect;
+            }
+        } else
+#endif
+        {
+            pdpe_addr = (env->nested_cr3 & ~0x1f) + ((gphys >> 27) & 0x18);
+            pdpe = x86_ldq_phys(cs, pdpe_addr);
+            if (!(pdpe & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            rsvd_mask |= PG_HI_USER_MASK;
+            if (pdpe & (rsvd_mask | PG_NX_MASK)) {
+                goto do_fault_rsvd;
+            }
+            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
+        }
+
+        pde_addr = (pdpe & PG_ADDRESS_MASK) + (((gphys >> 21) & 0x1ff) << 3);
+        pde = x86_ldq_phys(cs, pde_addr);
+        if (!(pde & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        if (pde & rsvd_mask) {
+            goto do_fault_rsvd;
+        }
+        ptep &= pde ^ PG_NX_MASK;
+        if (pde & PG_PSE_MASK) {
+            /* 2 MB page */
+            page_size = 2048 * 1024;
+            pte_addr = pde_addr;
+            pte = pde;
+            goto do_check_protect;
+        }
+        /* 4 KB page */
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            x86_stl_phys_notdirty(cs, pde_addr, pde);
+        }
+        pte_addr = (pde & PG_ADDRESS_MASK) + (((gphys >> 12) & 0x1ff) << 3);
+        pte = x86_ldq_phys(cs, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        if (pte & rsvd_mask) {
+            goto do_fault_rsvd;
+        }
+        /* combine pde and pte nx, user and rw protections */
+        ptep &= pte ^ PG_NX_MASK;
+        page_size = 4096;
+    } else {
+        uint32_t pde;
+
+        /* page directory entry */
+        pde_addr = (env->nested_cr3 & ~0xfff) + ((gphys >> 20) & 0xffc);
+        pde = x86_ldl_phys(cs, pde_addr);
+        if (!(pde & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        ptep = pde | PG_NX_MASK;
+
+        /* if host cr4 PSE bit is set, then we use a 4MB page */
+        if ((pde & PG_PSE_MASK) && (env->nested_pg_mode & SVM_NPT_PSE)) {
+            page_size = 4096 * 1024;
+            pte_addr = pde_addr;
+
+            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.
+             * Leave bits 20-13 in place for setting accessed/dirty bits below.
+             */
+            pte = pde | ((pde & 0x1fe000LL) << (32 - 13));
+            rsvd_mask = 0x200000;
+            goto do_check_protect_pse36;
+        }
+
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            x86_stl_phys_notdirty(cs, pde_addr, pde);
+        }
+
+        /* page directory entry */
+        pte_addr = (pde & ~0xfff) + ((gphys >> 10) & 0xffc);
+        pte = x86_ldl_phys(cs, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        /* combine pde and pte user and rw protections */
+        ptep &= pte | PG_NX_MASK;
+        page_size = 4096;
+        rsvd_mask = 0;
+    }
+
+ do_check_protect:
+    rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK;
+ do_check_protect_pse36:
+    if (pte & rsvd_mask) {
+        goto do_fault_rsvd;
+    }
+    ptep ^= PG_NX_MASK;
+
+    if (!(ptep & PG_USER_MASK)) {
+        goto do_fault_protect;
+    }
+    if (ptep & PG_NX_MASK) {
+        if (access_type == MMU_INST_FETCH) {
+            goto do_fault_protect;
+        }
+        *prot &= ~PAGE_EXEC;
+    }
+    if (!(ptep & PG_RW_MASK)) {
+        if (access_type == MMU_DATA_STORE) {
+            goto do_fault_protect;
+        }
+        *prot &= ~PAGE_WRITE;
+    }
+
+    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
+    page_offset = gphys & (page_size - 1);
+    return pte + page_offset;
+
+ do_fault_rsvd:
+    exit_info_1 |= SVM_NPTEXIT_RSVD;
+ do_fault_protect:
+    exit_info_1 |= SVM_NPTEXIT_P;
+ do_fault:
+    x86_stq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2),
+                 gphys);
+    exit_info_1 |= SVM_NPTEXIT_US;
+    if (access_type == MMU_DATA_STORE) {
+        exit_info_1 |= SVM_NPTEXIT_RW;
+    } else if (access_type == MMU_INST_FETCH) {
+        exit_info_1 |= SVM_NPTEXIT_ID;
+    }
+    if (prot) {
+        exit_info_1 |= SVM_NPTEXIT_GPA;
+    } else { /* page table access */
+        exit_info_1 |= SVM_NPTEXIT_GPT;
+    }
+    cpu_vmexit(env, SVM_EXIT_NPF, exit_info_1, env->retaddr);
+}
+
+/* return value:
+ * -1 = cannot handle fault
+ * 0  = nothing more to do
+ * 1  = generate PF fault
+ */
+static int handle_mmu_fault(CPUState *cs, vaddr addr, int size,
+                            int is_write1, int mmu_idx)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+    uint64_t ptep, pte;
+    int32_t a20_mask;
+    target_ulong pde_addr, pte_addr;
+    int error_code = 0;
+    int is_dirty, prot, page_size, is_write, is_user;
+    hwaddr paddr;
+    uint64_t rsvd_mask = PG_HI_RSVD_MASK;
+    uint32_t page_offset;
+    target_ulong vaddr;
+    uint32_t pkr;
+
+    is_user = mmu_idx == MMU_USER_IDX;
+#if defined(DEBUG_MMU)
+    printf("MMU fault: addr=%" VADDR_PRIx " w=%d u=%d eip=" TARGET_FMT_lx "\n",
+           addr, is_write1, is_user, env->eip);
+#endif
+    is_write = is_write1 & 1;
+
+    a20_mask = x86_get_a20_mask(env);
+    if (!(env->cr[0] & CR0_PG_MASK)) {
+        pte = addr;
+#ifdef TARGET_X86_64
+        if (!(env->hflags & HF_LMA_MASK)) {
+            /* Without long mode we can only address 32bits in real mode */
+            pte = (uint32_t)pte;
+        }
+#endif
+        prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
+        page_size = 4096;
+        goto do_mapping;
+    }
+
+    if (!(env->efer & MSR_EFER_NXE)) {
+        rsvd_mask |= PG_NX_MASK;
+    }
+
+    if (env->cr[4] & CR4_PAE_MASK) {
+        uint64_t pde, pdpe;
+        target_ulong pdpe_addr;
+
+#ifdef TARGET_X86_64
+        if (env->hflags & HF_LMA_MASK) {
+            bool la57 = env->cr[4] & CR4_LA57_MASK;
+            uint64_t pml5e_addr, pml5e;
+            uint64_t pml4e_addr, pml4e;
+            int32_t sext;
+
+            /* test virtual address sign extension */
+            sext = la57 ? (int64_t)addr >> 56 : (int64_t)addr >> 47;
+            if (sext != 0 && sext != -1) {
+                env->error_code = 0;
+                cs->exception_index = EXCP0D_GPF;
+                return 1;
+            }
+
+            if (la57) {
+                pml5e_addr = ((env->cr[3] & ~0xfff) +
+                        (((addr >> 48) & 0x1ff) << 3)) & a20_mask;
+                pml5e_addr = get_hphys(cs, pml5e_addr, MMU_DATA_STORE, NULL);
+                pml5e = x86_ldq_phys(cs, pml5e_addr);
+                if (!(pml5e & PG_PRESENT_MASK)) {
+                    goto do_fault;
+                }
+                if (pml5e & (rsvd_mask | PG_PSE_MASK)) {
+                    goto do_fault_rsvd;
+                }
+                if (!(pml5e & PG_ACCESSED_MASK)) {
+                    pml5e |= PG_ACCESSED_MASK;
+                    x86_stl_phys_notdirty(cs, pml5e_addr, pml5e);
+                }
+                ptep = pml5e ^ PG_NX_MASK;
+            } else {
+                pml5e = env->cr[3];
+                ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
+            }
+
+            pml4e_addr = ((pml5e & PG_ADDRESS_MASK) +
+                    (((addr >> 39) & 0x1ff) << 3)) & a20_mask;
+            pml4e_addr = get_hphys(cs, pml4e_addr, MMU_DATA_STORE, false);
+            pml4e = x86_ldq_phys(cs, pml4e_addr);
+            if (!(pml4e & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            if (pml4e & (rsvd_mask | PG_PSE_MASK)) {
+                goto do_fault_rsvd;
+            }
+            if (!(pml4e & PG_ACCESSED_MASK)) {
+                pml4e |= PG_ACCESSED_MASK;
+                x86_stl_phys_notdirty(cs, pml4e_addr, pml4e);
+            }
+            ptep &= pml4e ^ PG_NX_MASK;
+            pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) &
+                a20_mask;
+            pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, NULL);
+            pdpe = x86_ldq_phys(cs, pdpe_addr);
+            if (!(pdpe & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            if (pdpe & rsvd_mask) {
+                goto do_fault_rsvd;
+            }
+            ptep &= pdpe ^ PG_NX_MASK;
+            if (!(pdpe & PG_ACCESSED_MASK)) {
+                pdpe |= PG_ACCESSED_MASK;
+                x86_stl_phys_notdirty(cs, pdpe_addr, pdpe);
+            }
+            if (pdpe & PG_PSE_MASK) {
+                /* 1 GB page */
+                page_size = 1024 * 1024 * 1024;
+                pte_addr = pdpe_addr;
+                pte = pdpe;
+                goto do_check_protect;
+            }
+        } else
+#endif
+        {
+            /* XXX: load them when cr3 is loaded ? */
+            pdpe_addr = ((env->cr[3] & ~0x1f) + ((addr >> 27) & 0x18)) &
+                a20_mask;
+            pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, false);
+            pdpe = x86_ldq_phys(cs, pdpe_addr);
+            if (!(pdpe & PG_PRESENT_MASK)) {
+                goto do_fault;
+            }
+            rsvd_mask |= PG_HI_USER_MASK;
+            if (pdpe & (rsvd_mask | PG_NX_MASK)) {
+                goto do_fault_rsvd;
+            }
+            ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK;
+        }
+
+        pde_addr = ((pdpe & PG_ADDRESS_MASK) + (((addr >> 21) & 0x1ff) << 3)) &
+            a20_mask;
+        pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL);
+        pde = x86_ldq_phys(cs, pde_addr);
+        if (!(pde & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        if (pde & rsvd_mask) {
+            goto do_fault_rsvd;
+        }
+        ptep &= pde ^ PG_NX_MASK;
+        if (pde & PG_PSE_MASK) {
+            /* 2 MB page */
+            page_size = 2048 * 1024;
+            pte_addr = pde_addr;
+            pte = pde;
+            goto do_check_protect;
+        }
+        /* 4 KB page */
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            x86_stl_phys_notdirty(cs, pde_addr, pde);
+        }
+        pte_addr = ((pde & PG_ADDRESS_MASK) + (((addr >> 12) & 0x1ff) << 3)) &
+            a20_mask;
+        pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL);
+        pte = x86_ldq_phys(cs, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        if (pte & rsvd_mask) {
+            goto do_fault_rsvd;
+        }
+        /* combine pde and pte nx, user and rw protections */
+        ptep &= pte ^ PG_NX_MASK;
+        page_size = 4096;
+    } else {
+        uint32_t pde;
+
+        /* page directory entry */
+        pde_addr = ((env->cr[3] & ~0xfff) + ((addr >> 20) & 0xffc)) &
+            a20_mask;
+        pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL);
+        pde = x86_ldl_phys(cs, pde_addr);
+        if (!(pde & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        ptep = pde | PG_NX_MASK;
+
+        /* if PSE bit is set, then we use a 4MB page */
+        if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) {
+            page_size = 4096 * 1024;
+            pte_addr = pde_addr;
+
+            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.
+             * Leave bits 20-13 in place for setting accessed/dirty bits below.
+             */
+            pte = pde | ((pde & 0x1fe000LL) << (32 - 13));
+            rsvd_mask = 0x200000;
+            goto do_check_protect_pse36;
+        }
+
+        if (!(pde & PG_ACCESSED_MASK)) {
+            pde |= PG_ACCESSED_MASK;
+            x86_stl_phys_notdirty(cs, pde_addr, pde);
+        }
+
+        /* page directory entry */
+        pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) &
+            a20_mask;
+        pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL);
+        pte = x86_ldl_phys(cs, pte_addr);
+        if (!(pte & PG_PRESENT_MASK)) {
+            goto do_fault;
+        }
+        /* combine pde and pte user and rw protections */
+        ptep &= pte | PG_NX_MASK;
+        page_size = 4096;
+        rsvd_mask = 0;
+    }
+
+do_check_protect:
+    rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK;
+do_check_protect_pse36:
+    if (pte & rsvd_mask) {
+        goto do_fault_rsvd;
+    }
+    ptep ^= PG_NX_MASK;
+
+    /* can the page can be put in the TLB?  prot will tell us */
+    if (is_user && !(ptep & PG_USER_MASK)) {
+        goto do_fault_protect;
+    }
+
+    prot = 0;
+    if (mmu_idx != MMU_KSMAP_IDX || !(ptep & PG_USER_MASK)) {
+        prot |= PAGE_READ;
+        if ((ptep & PG_RW_MASK) || (!is_user && !(env->cr[0] & CR0_WP_MASK))) {
+            prot |= PAGE_WRITE;
+        }
+    }
+    if (!(ptep & PG_NX_MASK) &&
+        (mmu_idx == MMU_USER_IDX ||
+         !((env->cr[4] & CR4_SMEP_MASK) && (ptep & PG_USER_MASK)))) {
+        prot |= PAGE_EXEC;
+    }
+
+    if (!(env->hflags & HF_LMA_MASK)) {
+        pkr = 0;
+    } else if (ptep & PG_USER_MASK) {
+        pkr = env->cr[4] & CR4_PKE_MASK ? env->pkru : 0;
+    } else {
+        pkr = env->cr[4] & CR4_PKS_MASK ? env->pkrs : 0;
+    }
+    if (pkr) {
+        uint32_t pk = (pte & PG_PKRU_MASK) >> PG_PKRU_BIT;
+        uint32_t pkr_ad = (pkr >> pk * 2) & 1;
+        uint32_t pkr_wd = (pkr >> pk * 2) & 2;
+        uint32_t pkr_prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
+
+        if (pkr_ad) {
+            pkr_prot &= ~(PAGE_READ | PAGE_WRITE);
+        } else if (pkr_wd && (is_user || env->cr[0] & CR0_WP_MASK)) {
+            pkr_prot &= ~PAGE_WRITE;
+        }
+
+        prot &= pkr_prot;
+        if ((pkr_prot & (1 << is_write1)) == 0) {
+            assert(is_write1 != 2);
+            error_code |= PG_ERROR_PK_MASK;
+            goto do_fault_protect;
+        }
+    }
+
+    if ((prot & (1 << is_write1)) == 0) {
+        goto do_fault_protect;
+    }
+
+    /* yes, it can! */
+    is_dirty = is_write && !(pte & PG_DIRTY_MASK);
+    if (!(pte & PG_ACCESSED_MASK) || is_dirty) {
+        pte |= PG_ACCESSED_MASK;
+        if (is_dirty) {
+            pte |= PG_DIRTY_MASK;
+        }
+        x86_stl_phys_notdirty(cs, pte_addr, pte);
+    }
+
+    if (!(pte & PG_DIRTY_MASK)) {
+        /* only set write access if already dirty... otherwise wait
+           for dirty access */
+        assert(!is_write);
+        prot &= ~PAGE_WRITE;
+    }
+
+ do_mapping:
+    pte = pte & a20_mask;
+
+    /* align to page_size */
+    pte &= PG_ADDRESS_MASK & ~(page_size - 1);
+    page_offset = addr & (page_size - 1);
+    paddr = get_hphys(cs, pte + page_offset, is_write1, &prot);
+
+    /* Even if 4MB pages, we map only one 4KB page in the cache to
+       avoid filling it too fast */
+    vaddr = addr & TARGET_PAGE_MASK;
+    paddr &= TARGET_PAGE_MASK;
+
+    assert(prot & (1 << is_write1));
+    tlb_set_page_with_attrs(cs, vaddr, paddr, cpu_get_mem_attrs(env),
+                            prot, mmu_idx, page_size);
+    return 0;
+ do_fault_rsvd:
+    error_code |= PG_ERROR_RSVD_MASK;
+ do_fault_protect:
+    error_code |= PG_ERROR_P_MASK;
+ do_fault:
+    error_code |= (is_write << PG_ERROR_W_BIT);
+    if (is_user)
+        error_code |= PG_ERROR_U_MASK;
+    if (is_write1 == 2 &&
+        (((env->efer & MSR_EFER_NXE) &&
+          (env->cr[4] & CR4_PAE_MASK)) ||
+         (env->cr[4] & CR4_SMEP_MASK)))
+        error_code |= PG_ERROR_I_D_MASK;
+    if (env->intercept_exceptions & (1 << EXCP0E_PAGE)) {
+        /* cr2 is not modified in case of exceptions */
+        x86_stq_phys(cs,
+                 env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2),
+                 addr);
+    } else {
+        env->cr[2] = addr;
+    }
+    env->error_code = error_code;
+    cs->exception_index = EXCP0E_PAGE;
+    return 1;
+}
+
+bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size,
+                      MMUAccessType access_type, int mmu_idx,
+                      bool probe, uintptr_t retaddr)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    env->retaddr = retaddr;
+    if (handle_mmu_fault(cs, addr, size, access_type, mmu_idx)) {
+        /* FIXME: On error in get_hphys we have already jumped out.  */
+        g_assert(!probe);
+        raise_exception_err_ra(env, cs->exception_index,
+                               env->error_code, retaddr);
+    }
+    return true;
+}
diff --git a/target/i386/tcg/user/excp_helper.c b/target/i386/tcg/user/excp_helper.c
new file mode 100644
index 0000000000..a89b5228fd
--- /dev/null
+++ b/target/i386/tcg/user/excp_helper.c
@@ -0,0 +1,39 @@
+/*
+ *  x86 exception helpers - user-mode specific code
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/exec-all.h"
+#include "tcg/helper-tcg.h"
+
+bool x86_cpu_tlb_fill(CPUState *cs, vaddr addr, int size,
+                      MMUAccessType access_type, int mmu_idx,
+                      bool probe, uintptr_t retaddr)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    env->cr[2] = addr;
+    env->error_code = (access_type == MMU_DATA_STORE) << PG_ERROR_W_BIT;
+    env->error_code |= PG_ERROR_U_MASK;
+    cs->exception_index = EXCP0E_PAGE;
+    env->exception_is_int = 0;
+    env->exception_next_eip = -1;
+    cpu_loop_exit_restore(cs, retaddr);
+}
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index 35ba16dc3d..6d0a0a0fee 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -1,4 +1,5 @@
 i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'tcg-cpu.c',
   'smm_helper.c',
+  'excp_helper.c',
 ))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
index 7aecc53155..e0ef0f02e2 100644
--- a/target/i386/tcg/user/meson.build
+++ b/target/i386/tcg/user/meson.build
@@ -1,2 +1,3 @@
 i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
+  'excp_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 10/15] i386: split tcg btp_helper into softmmu and user parts
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (8 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 09/15] i386: split tcg excp_helper into softmmu and user parts Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 11:55   ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 11/15] i386: split misc helper into user and softmmu parts Claudio Fontana
                   ` (6 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/helper.h                 |   7 +
 target/i386/tcg/helper-tcg.h         |   3 +
 target/i386/tcg/bpt_helper.c         | 276 -------------------------
 target/i386/tcg/softmmu/bpt_helper.c | 293 +++++++++++++++++++++++++++
 target/i386/tcg/translate.c          |   4 +
 target/i386/tcg/softmmu/meson.build  |   1 +
 6 files changed, 308 insertions(+), 276 deletions(-)
 create mode 100644 target/i386/tcg/softmmu/bpt_helper.c

diff --git a/target/i386/helper.h b/target/i386/helper.h
index 8ffda4cdc6..095520f81f 100644
--- a/target/i386/helper.h
+++ b/target/i386/helper.h
@@ -46,7 +46,11 @@ DEF_HELPER_2(read_crN, tl, env, int)
 DEF_HELPER_3(write_crN, void, env, int, tl)
 DEF_HELPER_2(lmsw, void, env, tl)
 DEF_HELPER_1(clts, void, env)
+
+#ifndef CONFIG_USER_ONLY
 DEF_HELPER_FLAGS_3(set_dr, TCG_CALL_NO_WG, void, env, int, tl)
+#endif /* !CONFIG_USER_ONLY */
+
 DEF_HELPER_FLAGS_2(get_dr, TCG_CALL_NO_WG, tl, env, int)
 DEF_HELPER_2(invlpg, void, env, tl)
 
@@ -100,7 +104,10 @@ DEF_HELPER_3(outw, void, env, i32, i32)
 DEF_HELPER_2(inw, tl, env, i32)
 DEF_HELPER_3(outl, void, env, i32, i32)
 DEF_HELPER_2(inl, tl, env, i32)
+
+#ifndef CONFIG_USER_ONLY
 DEF_HELPER_FLAGS_4(bpt_io, TCG_CALL_NO_WG, void, env, i32, i32, tl)
+#endif /* !CONFIG_USER_ONLY */
 
 DEF_HELPER_3(svm_check_intercept_param, void, env, i32, i64)
 DEF_HELPER_4(svm_check_io, void, env, i32, i32, i32)
diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h
index c133c63555..b420b3356d 100644
--- a/target/i386/tcg/helper-tcg.h
+++ b/target/i386/tcg/helper-tcg.h
@@ -92,4 +92,7 @@ void do_interrupt_x86_hardirq(CPUX86State *env, int intno, int is_hw);
 /* smm_helper.c */
 void do_smm_enter(X86CPU *cpu);
 
+/* bpt_helper.c */
+bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update);
+
 #endif /* I386_HELPER_TCG_H */
diff --git a/target/i386/tcg/bpt_helper.c b/target/i386/tcg/bpt_helper.c
index 979230ac12..fb2a65ac9c 100644
--- a/target/i386/tcg/bpt_helper.c
+++ b/target/i386/tcg/bpt_helper.c
@@ -19,223 +19,9 @@
 
 #include "qemu/osdep.h"
 #include "cpu.h"
-#include "exec/exec-all.h"
 #include "exec/helper-proto.h"
 #include "helper-tcg.h"
 
-
-#ifndef CONFIG_USER_ONLY
-static inline bool hw_local_breakpoint_enabled(unsigned long dr7, int index)
-{
-    return (dr7 >> (index * 2)) & 1;
-}
-
-static inline bool hw_global_breakpoint_enabled(unsigned long dr7, int index)
-{
-    return (dr7 >> (index * 2)) & 2;
-
-}
-static inline bool hw_breakpoint_enabled(unsigned long dr7, int index)
-{
-    return hw_global_breakpoint_enabled(dr7, index) ||
-           hw_local_breakpoint_enabled(dr7, index);
-}
-
-static inline int hw_breakpoint_type(unsigned long dr7, int index)
-{
-    return (dr7 >> (DR7_TYPE_SHIFT + (index * 4))) & 3;
-}
-
-static inline int hw_breakpoint_len(unsigned long dr7, int index)
-{
-    int len = ((dr7 >> (DR7_LEN_SHIFT + (index * 4))) & 3);
-    return (len == 2) ? 8 : len + 1;
-}
-
-static int hw_breakpoint_insert(CPUX86State *env, int index)
-{
-    CPUState *cs = env_cpu(env);
-    target_ulong dr7 = env->dr[7];
-    target_ulong drN = env->dr[index];
-    int err = 0;
-
-    switch (hw_breakpoint_type(dr7, index)) {
-    case DR7_TYPE_BP_INST:
-        if (hw_breakpoint_enabled(dr7, index)) {
-            err = cpu_breakpoint_insert(cs, drN, BP_CPU,
-                                        &env->cpu_breakpoint[index]);
-        }
-        break;
-
-    case DR7_TYPE_IO_RW:
-        /* Notice when we should enable calls to bpt_io.  */
-        return hw_breakpoint_enabled(env->dr[7], index)
-               ? HF_IOBPT_MASK : 0;
-
-    case DR7_TYPE_DATA_WR:
-        if (hw_breakpoint_enabled(dr7, index)) {
-            err = cpu_watchpoint_insert(cs, drN,
-                                        hw_breakpoint_len(dr7, index),
-                                        BP_CPU | BP_MEM_WRITE,
-                                        &env->cpu_watchpoint[index]);
-        }
-        break;
-
-    case DR7_TYPE_DATA_RW:
-        if (hw_breakpoint_enabled(dr7, index)) {
-            err = cpu_watchpoint_insert(cs, drN,
-                                        hw_breakpoint_len(dr7, index),
-                                        BP_CPU | BP_MEM_ACCESS,
-                                        &env->cpu_watchpoint[index]);
-        }
-        break;
-    }
-    if (err) {
-        env->cpu_breakpoint[index] = NULL;
-    }
-    return 0;
-}
-
-static void hw_breakpoint_remove(CPUX86State *env, int index)
-{
-    CPUState *cs = env_cpu(env);
-
-    switch (hw_breakpoint_type(env->dr[7], index)) {
-    case DR7_TYPE_BP_INST:
-        if (env->cpu_breakpoint[index]) {
-            cpu_breakpoint_remove_by_ref(cs, env->cpu_breakpoint[index]);
-            env->cpu_breakpoint[index] = NULL;
-        }
-        break;
-
-    case DR7_TYPE_DATA_WR:
-    case DR7_TYPE_DATA_RW:
-        if (env->cpu_breakpoint[index]) {
-            cpu_watchpoint_remove_by_ref(cs, env->cpu_watchpoint[index]);
-            env->cpu_breakpoint[index] = NULL;
-        }
-        break;
-
-    case DR7_TYPE_IO_RW:
-        /* HF_IOBPT_MASK cleared elsewhere.  */
-        break;
-    }
-}
-
-void cpu_x86_update_dr7(CPUX86State *env, uint32_t new_dr7)
-{
-    target_ulong old_dr7 = env->dr[7];
-    int iobpt = 0;
-    int i;
-
-    new_dr7 |= DR7_FIXED_1;
-
-    /* If nothing is changing except the global/local enable bits,
-       then we can make the change more efficient.  */
-    if (((old_dr7 ^ new_dr7) & ~0xff) == 0) {
-        /* Fold the global and local enable bits together into the
-           global fields, then xor to show which registers have
-           changed collective enable state.  */
-        int mod = ((old_dr7 | old_dr7 * 2) ^ (new_dr7 | new_dr7 * 2)) & 0xff;
-
-        for (i = 0; i < DR7_MAX_BP; i++) {
-            if ((mod & (2 << i * 2)) && !hw_breakpoint_enabled(new_dr7, i)) {
-                hw_breakpoint_remove(env, i);
-            }
-        }
-        env->dr[7] = new_dr7;
-        for (i = 0; i < DR7_MAX_BP; i++) {
-            if (mod & (2 << i * 2) && hw_breakpoint_enabled(new_dr7, i)) {
-                iobpt |= hw_breakpoint_insert(env, i);
-            } else if (hw_breakpoint_type(new_dr7, i) == DR7_TYPE_IO_RW
-                       && hw_breakpoint_enabled(new_dr7, i)) {
-                iobpt |= HF_IOBPT_MASK;
-            }
-        }
-    } else {
-        for (i = 0; i < DR7_MAX_BP; i++) {
-            hw_breakpoint_remove(env, i);
-        }
-        env->dr[7] = new_dr7;
-        for (i = 0; i < DR7_MAX_BP; i++) {
-            iobpt |= hw_breakpoint_insert(env, i);
-        }
-    }
-
-    env->hflags = (env->hflags & ~HF_IOBPT_MASK) | iobpt;
-}
-
-static bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update)
-{
-    target_ulong dr6;
-    int reg;
-    bool hit_enabled = false;
-
-    dr6 = env->dr[6] & ~0xf;
-    for (reg = 0; reg < DR7_MAX_BP; reg++) {
-        bool bp_match = false;
-        bool wp_match = false;
-
-        switch (hw_breakpoint_type(env->dr[7], reg)) {
-        case DR7_TYPE_BP_INST:
-            if (env->dr[reg] == env->eip) {
-                bp_match = true;
-            }
-            break;
-        case DR7_TYPE_DATA_WR:
-        case DR7_TYPE_DATA_RW:
-            if (env->cpu_watchpoint[reg] &&
-                env->cpu_watchpoint[reg]->flags & BP_WATCHPOINT_HIT) {
-                wp_match = true;
-            }
-            break;
-        case DR7_TYPE_IO_RW:
-            break;
-        }
-        if (bp_match || wp_match) {
-            dr6 |= 1 << reg;
-            if (hw_breakpoint_enabled(env->dr[7], reg)) {
-                hit_enabled = true;
-            }
-        }
-    }
-
-    if (hit_enabled || force_dr6_update) {
-        env->dr[6] = dr6;
-    }
-
-    return hit_enabled;
-}
-
-void breakpoint_handler(CPUState *cs)
-{
-    X86CPU *cpu = X86_CPU(cs);
-    CPUX86State *env = &cpu->env;
-    CPUBreakpoint *bp;
-
-    if (cs->watchpoint_hit) {
-        if (cs->watchpoint_hit->flags & BP_CPU) {
-            cs->watchpoint_hit = NULL;
-            if (check_hw_breakpoints(env, false)) {
-                raise_exception(env, EXCP01_DB);
-            } else {
-                cpu_loop_exit_noexc(cs);
-            }
-        }
-    } else {
-        QTAILQ_FOREACH(bp, &cs->breakpoints, entry) {
-            if (bp->pc == env->eip) {
-                if (bp->flags & BP_CPU) {
-                    check_hw_breakpoints(env, true);
-                    raise_exception(env, EXCP01_DB);
-                }
-                break;
-            }
-        }
-    }
-}
-#endif
-
 void helper_single_step(CPUX86State *env)
 {
 #ifndef CONFIG_USER_ONLY
@@ -252,41 +38,6 @@ void helper_rechecking_single_step(CPUX86State *env)
     }
 }
 
-void helper_set_dr(CPUX86State *env, int reg, target_ulong t0)
-{
-#ifndef CONFIG_USER_ONLY
-    switch (reg) {
-    case 0: case 1: case 2: case 3:
-        if (hw_breakpoint_enabled(env->dr[7], reg)
-            && hw_breakpoint_type(env->dr[7], reg) != DR7_TYPE_IO_RW) {
-            hw_breakpoint_remove(env, reg);
-            env->dr[reg] = t0;
-            hw_breakpoint_insert(env, reg);
-        } else {
-            env->dr[reg] = t0;
-        }
-        return;
-    case 4:
-        if (env->cr[4] & CR4_DE_MASK) {
-            break;
-        }
-        /* fallthru */
-    case 6:
-        env->dr[6] = t0 | DR6_FIXED_1;
-        return;
-    case 5:
-        if (env->cr[4] & CR4_DE_MASK) {
-            break;
-        }
-        /* fallthru */
-    case 7:
-        cpu_x86_update_dr7(env, t0);
-        return;
-    }
-    raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
-#endif
-}
-
 target_ulong helper_get_dr(CPUX86State *env, int reg)
 {
     switch (reg) {
@@ -307,30 +58,3 @@ target_ulong helper_get_dr(CPUX86State *env, int reg)
     }
     raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
 }
-
-/* Check if Port I/O is trapped by a breakpoint.  */
-void helper_bpt_io(CPUX86State *env, uint32_t port,
-                   uint32_t size, target_ulong next_eip)
-{
-#ifndef CONFIG_USER_ONLY
-    target_ulong dr7 = env->dr[7];
-    int i, hit = 0;
-
-    for (i = 0; i < DR7_MAX_BP; ++i) {
-        if (hw_breakpoint_type(dr7, i) == DR7_TYPE_IO_RW
-            && hw_breakpoint_enabled(dr7, i)) {
-            int bpt_len = hw_breakpoint_len(dr7, i);
-            if (port + size - 1 >= env->dr[i]
-                && port <= env->dr[i] + bpt_len - 1) {
-                hit |= 1 << i;
-            }
-        }
-    }
-
-    if (hit) {
-        env->dr[6] = (env->dr[6] & ~0xf) | hit;
-        env->eip = next_eip;
-        raise_exception(env, EXCP01_DB);
-    }
-#endif
-}
diff --git a/target/i386/tcg/softmmu/bpt_helper.c b/target/i386/tcg/softmmu/bpt_helper.c
new file mode 100644
index 0000000000..1f3dd09cad
--- /dev/null
+++ b/target/i386/tcg/softmmu/bpt_helper.c
@@ -0,0 +1,293 @@
+/*
+ *  i386 breakpoint helpers - softmmu-only code
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/exec-all.h"
+#include "exec/helper-proto.h"
+#include "tcg/helper-tcg.h"
+
+
+static inline bool hw_local_breakpoint_enabled(unsigned long dr7, int index)
+{
+    return (dr7 >> (index * 2)) & 1;
+}
+
+static inline bool hw_global_breakpoint_enabled(unsigned long dr7, int index)
+{
+    return (dr7 >> (index * 2)) & 2;
+
+}
+static inline bool hw_breakpoint_enabled(unsigned long dr7, int index)
+{
+    return hw_global_breakpoint_enabled(dr7, index) ||
+           hw_local_breakpoint_enabled(dr7, index);
+}
+
+static inline int hw_breakpoint_type(unsigned long dr7, int index)
+{
+    return (dr7 >> (DR7_TYPE_SHIFT + (index * 4))) & 3;
+}
+
+static inline int hw_breakpoint_len(unsigned long dr7, int index)
+{
+    int len = ((dr7 >> (DR7_LEN_SHIFT + (index * 4))) & 3);
+    return (len == 2) ? 8 : len + 1;
+}
+
+static int hw_breakpoint_insert(CPUX86State *env, int index)
+{
+    CPUState *cs = env_cpu(env);
+    target_ulong dr7 = env->dr[7];
+    target_ulong drN = env->dr[index];
+    int err = 0;
+
+    switch (hw_breakpoint_type(dr7, index)) {
+    case DR7_TYPE_BP_INST:
+        if (hw_breakpoint_enabled(dr7, index)) {
+            err = cpu_breakpoint_insert(cs, drN, BP_CPU,
+                                        &env->cpu_breakpoint[index]);
+        }
+        break;
+
+    case DR7_TYPE_IO_RW:
+        /* Notice when we should enable calls to bpt_io.  */
+        return hw_breakpoint_enabled(env->dr[7], index)
+               ? HF_IOBPT_MASK : 0;
+
+    case DR7_TYPE_DATA_WR:
+        if (hw_breakpoint_enabled(dr7, index)) {
+            err = cpu_watchpoint_insert(cs, drN,
+                                        hw_breakpoint_len(dr7, index),
+                                        BP_CPU | BP_MEM_WRITE,
+                                        &env->cpu_watchpoint[index]);
+        }
+        break;
+
+    case DR7_TYPE_DATA_RW:
+        if (hw_breakpoint_enabled(dr7, index)) {
+            err = cpu_watchpoint_insert(cs, drN,
+                                        hw_breakpoint_len(dr7, index),
+                                        BP_CPU | BP_MEM_ACCESS,
+                                        &env->cpu_watchpoint[index]);
+        }
+        break;
+    }
+    if (err) {
+        env->cpu_breakpoint[index] = NULL;
+    }
+    return 0;
+}
+
+static void hw_breakpoint_remove(CPUX86State *env, int index)
+{
+    CPUState *cs = env_cpu(env);
+
+    switch (hw_breakpoint_type(env->dr[7], index)) {
+    case DR7_TYPE_BP_INST:
+        if (env->cpu_breakpoint[index]) {
+            cpu_breakpoint_remove_by_ref(cs, env->cpu_breakpoint[index]);
+            env->cpu_breakpoint[index] = NULL;
+        }
+        break;
+
+    case DR7_TYPE_DATA_WR:
+    case DR7_TYPE_DATA_RW:
+        if (env->cpu_breakpoint[index]) {
+            cpu_watchpoint_remove_by_ref(cs, env->cpu_watchpoint[index]);
+            env->cpu_breakpoint[index] = NULL;
+        }
+        break;
+
+    case DR7_TYPE_IO_RW:
+        /* HF_IOBPT_MASK cleared elsewhere.  */
+        break;
+    }
+}
+
+void cpu_x86_update_dr7(CPUX86State *env, uint32_t new_dr7)
+{
+    target_ulong old_dr7 = env->dr[7];
+    int iobpt = 0;
+    int i;
+
+    new_dr7 |= DR7_FIXED_1;
+
+    /* If nothing is changing except the global/local enable bits,
+       then we can make the change more efficient.  */
+    if (((old_dr7 ^ new_dr7) & ~0xff) == 0) {
+        /* Fold the global and local enable bits together into the
+           global fields, then xor to show which registers have
+           changed collective enable state.  */
+        int mod = ((old_dr7 | old_dr7 * 2) ^ (new_dr7 | new_dr7 * 2)) & 0xff;
+
+        for (i = 0; i < DR7_MAX_BP; i++) {
+            if ((mod & (2 << i * 2)) && !hw_breakpoint_enabled(new_dr7, i)) {
+                hw_breakpoint_remove(env, i);
+            }
+        }
+        env->dr[7] = new_dr7;
+        for (i = 0; i < DR7_MAX_BP; i++) {
+            if (mod & (2 << i * 2) && hw_breakpoint_enabled(new_dr7, i)) {
+                iobpt |= hw_breakpoint_insert(env, i);
+            } else if (hw_breakpoint_type(new_dr7, i) == DR7_TYPE_IO_RW
+                       && hw_breakpoint_enabled(new_dr7, i)) {
+                iobpt |= HF_IOBPT_MASK;
+            }
+        }
+    } else {
+        for (i = 0; i < DR7_MAX_BP; i++) {
+            hw_breakpoint_remove(env, i);
+        }
+        env->dr[7] = new_dr7;
+        for (i = 0; i < DR7_MAX_BP; i++) {
+            iobpt |= hw_breakpoint_insert(env, i);
+        }
+    }
+
+    env->hflags = (env->hflags & ~HF_IOBPT_MASK) | iobpt;
+}
+
+bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update)
+{
+    target_ulong dr6;
+    int reg;
+    bool hit_enabled = false;
+
+    dr6 = env->dr[6] & ~0xf;
+    for (reg = 0; reg < DR7_MAX_BP; reg++) {
+        bool bp_match = false;
+        bool wp_match = false;
+
+        switch (hw_breakpoint_type(env->dr[7], reg)) {
+        case DR7_TYPE_BP_INST:
+            if (env->dr[reg] == env->eip) {
+                bp_match = true;
+            }
+            break;
+        case DR7_TYPE_DATA_WR:
+        case DR7_TYPE_DATA_RW:
+            if (env->cpu_watchpoint[reg] &&
+                env->cpu_watchpoint[reg]->flags & BP_WATCHPOINT_HIT) {
+                wp_match = true;
+            }
+            break;
+        case DR7_TYPE_IO_RW:
+            break;
+        }
+        if (bp_match || wp_match) {
+            dr6 |= 1 << reg;
+            if (hw_breakpoint_enabled(env->dr[7], reg)) {
+                hit_enabled = true;
+            }
+        }
+    }
+
+    if (hit_enabled || force_dr6_update) {
+        env->dr[6] = dr6;
+    }
+
+    return hit_enabled;
+}
+
+void breakpoint_handler(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+    CPUBreakpoint *bp;
+
+    if (cs->watchpoint_hit) {
+        if (cs->watchpoint_hit->flags & BP_CPU) {
+            cs->watchpoint_hit = NULL;
+            if (check_hw_breakpoints(env, false)) {
+                raise_exception(env, EXCP01_DB);
+            } else {
+                cpu_loop_exit_noexc(cs);
+            }
+        }
+    } else {
+        QTAILQ_FOREACH(bp, &cs->breakpoints, entry) {
+            if (bp->pc == env->eip) {
+                if (bp->flags & BP_CPU) {
+                    check_hw_breakpoints(env, true);
+                    raise_exception(env, EXCP01_DB);
+                }
+                break;
+            }
+        }
+    }
+}
+
+void helper_set_dr(CPUX86State *env, int reg, target_ulong t0)
+{
+    switch (reg) {
+    case 0: case 1: case 2: case 3:
+        if (hw_breakpoint_enabled(env->dr[7], reg)
+            && hw_breakpoint_type(env->dr[7], reg) != DR7_TYPE_IO_RW) {
+            hw_breakpoint_remove(env, reg);
+            env->dr[reg] = t0;
+            hw_breakpoint_insert(env, reg);
+        } else {
+            env->dr[reg] = t0;
+        }
+        return;
+    case 4:
+        if (env->cr[4] & CR4_DE_MASK) {
+            break;
+        }
+        /* fallthru */
+    case 6:
+        env->dr[6] = t0 | DR6_FIXED_1;
+        return;
+    case 5:
+        if (env->cr[4] & CR4_DE_MASK) {
+            break;
+        }
+        /* fallthru */
+    case 7:
+        cpu_x86_update_dr7(env, t0);
+        return;
+    }
+    raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
+}
+
+/* Check if Port I/O is trapped by a breakpoint.  */
+void helper_bpt_io(CPUX86State *env, uint32_t port,
+                   uint32_t size, target_ulong next_eip)
+{
+    target_ulong dr7 = env->dr[7];
+    int i, hit = 0;
+
+    for (i = 0; i < DR7_MAX_BP; ++i) {
+        if (hw_breakpoint_type(dr7, i) == DR7_TYPE_IO_RW
+            && hw_breakpoint_enabled(dr7, i)) {
+            int bpt_len = hw_breakpoint_len(dr7, i);
+            if (port + size - 1 >= env->dr[i]
+                && port <= env->dr[i] + bpt_len - 1) {
+                hit |= 1 << i;
+            }
+        }
+    }
+
+    if (hit) {
+        env->dr[6] = (env->dr[6] & ~0xf) | hit;
+        env->eip = next_eip;
+        raise_exception(env, EXCP01_DB);
+    }
+}
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index 5075ac4830..35f5d5c73e 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -1116,6 +1116,7 @@ static inline void gen_cmps(DisasContext *s, MemOp ot)
 
 static void gen_bpt_io(DisasContext *s, TCGv_i32 t_port, int ot)
 {
+#ifndef CONFIG_USER_ONLY
     if (s->flags & HF_IOBPT_MASK) {
         TCGv_i32 t_size = tcg_const_i32(1 << ot);
         TCGv t_next = tcg_const_tl(s->pc - s->cs_base);
@@ -1124,6 +1125,7 @@ static void gen_bpt_io(DisasContext *s, TCGv_i32 t_port, int ot)
         tcg_temp_free_i32(t_size);
         tcg_temp_free(t_next);
     }
+#endif /* !CONFIG_USER_ONLY */
 }
 
 
@@ -8074,7 +8076,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
                 gen_svm_check_intercept(s, pc_start, SVM_EXIT_WRITE_DR0 + reg);
                 gen_op_mov_v_reg(s, ot, s->T0, rm);
                 tcg_gen_movi_i32(s->tmp2_i32, reg);
+#ifndef CONFIG_USER_ONLY
                 gen_helper_set_dr(cpu_env, s->tmp2_i32, s->T0);
+#endif /* CONFIG_USER_ONLY */
                 gen_jmp_im(s, s->pc - s->cs_base);
                 gen_eob(s);
             } else {
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index 6d0a0a0fee..1580950141 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -2,4 +2,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'tcg-cpu.c',
   'smm_helper.c',
   'excp_helper.c',
+  'bpt_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 11/15] i386: split misc helper into user and softmmu parts
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (9 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 10/15] i386: split tcg btp_helper " Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 12/15] i386: separate fpu_helper " Claudio Fontana
                   ` (5 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/tcg/misc_helper.c         | 463 --------------------------
 target/i386/tcg/softmmu/misc_helper.c | 438 ++++++++++++++++++++++++
 target/i386/tcg/user/misc_helper.c    |  72 ++++
 target/i386/tcg/softmmu/meson.build   |   1 +
 target/i386/tcg/user/meson.build      |   1 +
 5 files changed, 512 insertions(+), 463 deletions(-)
 create mode 100644 target/i386/tcg/softmmu/misc_helper.c
 create mode 100644 target/i386/tcg/user/misc_helper.c

diff --git a/target/i386/tcg/misc_helper.c b/target/i386/tcg/misc_helper.c
index f02e4fd400..82fb7037ac 100644
--- a/target/i386/tcg/misc_helper.c
+++ b/target/i386/tcg/misc_helper.c
@@ -18,12 +18,9 @@
  */
 
 #include "qemu/osdep.h"
-#include "qemu/main-loop.h"
 #include "cpu.h"
 #include "exec/helper-proto.h"
 #include "exec/exec-all.h"
-#include "exec/cpu_ldst.h"
-#include "exec/address-spaces.h"
 #include "helper-tcg.h"
 
 /*
@@ -39,69 +36,6 @@ void cpu_load_eflags(CPUX86State *env, int eflags, int update_mask)
         (eflags & update_mask) | 0x2;
 }
 
-void helper_outb(CPUX86State *env, uint32_t port, uint32_t data)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "outb: port=0x%04x, data=%02x\n", port, data);
-#else
-    address_space_stb(&address_space_io, port, data,
-                      cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
-target_ulong helper_inb(CPUX86State *env, uint32_t port)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "inb: port=0x%04x\n", port);
-    return 0;
-#else
-    return address_space_ldub(&address_space_io, port,
-                              cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
-void helper_outw(CPUX86State *env, uint32_t port, uint32_t data)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "outw: port=0x%04x, data=%04x\n", port, data);
-#else
-    address_space_stw(&address_space_io, port, data,
-                      cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
-target_ulong helper_inw(CPUX86State *env, uint32_t port)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "inw: port=0x%04x\n", port);
-    return 0;
-#else
-    return address_space_lduw(&address_space_io, port,
-                              cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
-void helper_outl(CPUX86State *env, uint32_t port, uint32_t data)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "outl: port=0x%04x, data=%08x\n", port, data);
-#else
-    address_space_stl(&address_space_io, port, data,
-                      cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
-target_ulong helper_inl(CPUX86State *env, uint32_t port)
-{
-#ifdef CONFIG_USER_ONLY
-    fprintf(stderr, "inl: port=0x%04x\n", port);
-    return 0;
-#else
-    return address_space_ldl(&address_space_io, port,
-                             cpu_get_mem_attrs(env), NULL);
-#endif
-}
-
 void helper_into(CPUX86State *env, int next_eip_addend)
 {
     int eflags;
@@ -126,64 +60,6 @@ void helper_cpuid(CPUX86State *env)
     env->regs[R_EDX] = edx;
 }
 
-#if defined(CONFIG_USER_ONLY)
-target_ulong helper_read_crN(CPUX86State *env, int reg)
-{
-    return 0;
-}
-
-void helper_write_crN(CPUX86State *env, int reg, target_ulong t0)
-{
-}
-#else
-target_ulong helper_read_crN(CPUX86State *env, int reg)
-{
-    target_ulong val;
-
-    cpu_svm_check_intercept_param(env, SVM_EXIT_READ_CR0 + reg, 0, GETPC());
-    switch (reg) {
-    default:
-        val = env->cr[reg];
-        break;
-    case 8:
-        if (!(env->hflags2 & HF2_VINTR_MASK)) {
-            val = cpu_get_apic_tpr(env_archcpu(env)->apic_state);
-        } else {
-            val = env->v_tpr;
-        }
-        break;
-    }
-    return val;
-}
-
-void helper_write_crN(CPUX86State *env, int reg, target_ulong t0)
-{
-    cpu_svm_check_intercept_param(env, SVM_EXIT_WRITE_CR0 + reg, 0, GETPC());
-    switch (reg) {
-    case 0:
-        cpu_x86_update_cr0(env, t0);
-        break;
-    case 3:
-        cpu_x86_update_cr3(env, t0);
-        break;
-    case 4:
-        cpu_x86_update_cr4(env, t0);
-        break;
-    case 8:
-        if (!(env->hflags2 & HF2_VINTR_MASK)) {
-            qemu_mutex_lock_iothread();
-            cpu_set_apic_tpr(env_archcpu(env)->apic_state, t0);
-            qemu_mutex_unlock_iothread();
-        }
-        env->v_tpr = t0 & 0x0f;
-        break;
-    default:
-        env->cr[reg] = t0;
-        break;
-    }
-}
-#endif
-
 void helper_lmsw(CPUX86State *env, target_ulong t0)
 {
     /* only 4 lower bits of CR0 are modified. PE cannot be set to zero
@@ -232,345 +108,6 @@ void helper_rdpmc(CPUX86State *env)
     raise_exception_err(env, EXCP06_ILLOP, 0);
 }
 
-#if defined(CONFIG_USER_ONLY)
-void helper_wrmsr(CPUX86State *env)
-{
-}
-
-void helper_rdmsr(CPUX86State *env)
-{
-}
-#else
-void helper_wrmsr(CPUX86State *env)
-{
-    uint64_t val;
-    CPUState *cs = env_cpu(env);
-
-    cpu_svm_check_intercept_param(env, SVM_EXIT_MSR, 1, GETPC());
-
-    val = ((uint32_t)env->regs[R_EAX]) |
-        ((uint64_t)((uint32_t)env->regs[R_EDX]) << 32);
-
-    switch ((uint32_t)env->regs[R_ECX]) {
-    case MSR_IA32_SYSENTER_CS:
-        env->sysenter_cs = val & 0xffff;
-        break;
-    case MSR_IA32_SYSENTER_ESP:
-        env->sysenter_esp = val;
-        break;
-    case MSR_IA32_SYSENTER_EIP:
-        env->sysenter_eip = val;
-        break;
-    case MSR_IA32_APICBASE:
-        cpu_set_apic_base(env_archcpu(env)->apic_state, val);
-        break;
-    case MSR_EFER:
-        {
-            uint64_t update_mask;
-
-            update_mask = 0;
-            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_SYSCALL) {
-                update_mask |= MSR_EFER_SCE;
-            }
-            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM) {
-                update_mask |= MSR_EFER_LME;
-            }
-            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_FFXSR) {
-                update_mask |= MSR_EFER_FFXSR;
-            }
-            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_NX) {
-                update_mask |= MSR_EFER_NXE;
-            }
-            if (env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_SVM) {
-                update_mask |= MSR_EFER_SVME;
-            }
-            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_FFXSR) {
-                update_mask |= MSR_EFER_FFXSR;
-            }
-            cpu_load_efer(env, (env->efer & ~update_mask) |
-                          (val & update_mask));
-        }
-        break;
-    case MSR_STAR:
-        env->star = val;
-        break;
-    case MSR_PAT:
-        env->pat = val;
-        break;
-    case MSR_IA32_PKRS:
-        if (val & 0xFFFFFFFF00000000ull) {
-            goto error;
-        }
-        env->pkrs = val;
-        tlb_flush(cs);
-        break;
-    case MSR_VM_HSAVE_PA:
-        env->vm_hsave = val;
-        break;
-#ifdef TARGET_X86_64
-    case MSR_LSTAR:
-        env->lstar = val;
-        break;
-    case MSR_CSTAR:
-        env->cstar = val;
-        break;
-    case MSR_FMASK:
-        env->fmask = val;
-        break;
-    case MSR_FSBASE:
-        env->segs[R_FS].base = val;
-        break;
-    case MSR_GSBASE:
-        env->segs[R_GS].base = val;
-        break;
-    case MSR_KERNELGSBASE:
-        env->kernelgsbase = val;
-        break;
-#endif
-    case MSR_MTRRphysBase(0):
-    case MSR_MTRRphysBase(1):
-    case MSR_MTRRphysBase(2):
-    case MSR_MTRRphysBase(3):
-    case MSR_MTRRphysBase(4):
-    case MSR_MTRRphysBase(5):
-    case MSR_MTRRphysBase(6):
-    case MSR_MTRRphysBase(7):
-        env->mtrr_var[((uint32_t)env->regs[R_ECX] -
-                       MSR_MTRRphysBase(0)) / 2].base = val;
-        break;
-    case MSR_MTRRphysMask(0):
-    case MSR_MTRRphysMask(1):
-    case MSR_MTRRphysMask(2):
-    case MSR_MTRRphysMask(3):
-    case MSR_MTRRphysMask(4):
-    case MSR_MTRRphysMask(5):
-    case MSR_MTRRphysMask(6):
-    case MSR_MTRRphysMask(7):
-        env->mtrr_var[((uint32_t)env->regs[R_ECX] -
-                       MSR_MTRRphysMask(0)) / 2].mask = val;
-        break;
-    case MSR_MTRRfix64K_00000:
-        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
-                        MSR_MTRRfix64K_00000] = val;
-        break;
-    case MSR_MTRRfix16K_80000:
-    case MSR_MTRRfix16K_A0000:
-        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
-                        MSR_MTRRfix16K_80000 + 1] = val;
-        break;
-    case MSR_MTRRfix4K_C0000:
-    case MSR_MTRRfix4K_C8000:
-    case MSR_MTRRfix4K_D0000:
-    case MSR_MTRRfix4K_D8000:
-    case MSR_MTRRfix4K_E0000:
-    case MSR_MTRRfix4K_E8000:
-    case MSR_MTRRfix4K_F0000:
-    case MSR_MTRRfix4K_F8000:
-        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
-                        MSR_MTRRfix4K_C0000 + 3] = val;
-        break;
-    case MSR_MTRRdefType:
-        env->mtrr_deftype = val;
-        break;
-    case MSR_MCG_STATUS:
-        env->mcg_status = val;
-        break;
-    case MSR_MCG_CTL:
-        if ((env->mcg_cap & MCG_CTL_P)
-            && (val == 0 || val == ~(uint64_t)0)) {
-            env->mcg_ctl = val;
-        }
-        break;
-    case MSR_TSC_AUX:
-        env->tsc_aux = val;
-        break;
-    case MSR_IA32_MISC_ENABLE:
-        env->msr_ia32_misc_enable = val;
-        break;
-    case MSR_IA32_BNDCFGS:
-        /* FIXME: #GP if reserved bits are set.  */
-        /* FIXME: Extend highest implemented bit of linear address.  */
-        env->msr_bndcfgs = val;
-        cpu_sync_bndcs_hflags(env);
-        break;
-    default:
-        if ((uint32_t)env->regs[R_ECX] >= MSR_MC0_CTL
-            && (uint32_t)env->regs[R_ECX] < MSR_MC0_CTL +
-            (4 * env->mcg_cap & 0xff)) {
-            uint32_t offset = (uint32_t)env->regs[R_ECX] - MSR_MC0_CTL;
-            if ((offset & 0x3) != 0
-                || (val == 0 || val == ~(uint64_t)0)) {
-                env->mce_banks[offset] = val;
-            }
-            break;
-        }
-        /* XXX: exception? */
-        break;
-    }
-    return;
-error:
-    raise_exception_err_ra(env, EXCP0D_GPF, 0, GETPC());
-}
-
-void helper_rdmsr(CPUX86State *env)
-{
-    X86CPU *x86_cpu = env_archcpu(env);
-    uint64_t val;
-
-    cpu_svm_check_intercept_param(env, SVM_EXIT_MSR, 0, GETPC());
-
-    switch ((uint32_t)env->regs[R_ECX]) {
-    case MSR_IA32_SYSENTER_CS:
-        val = env->sysenter_cs;
-        break;
-    case MSR_IA32_SYSENTER_ESP:
-        val = env->sysenter_esp;
-        break;
-    case MSR_IA32_SYSENTER_EIP:
-        val = env->sysenter_eip;
-        break;
-    case MSR_IA32_APICBASE:
-        val = cpu_get_apic_base(env_archcpu(env)->apic_state);
-        break;
-    case MSR_EFER:
-        val = env->efer;
-        break;
-    case MSR_STAR:
-        val = env->star;
-        break;
-    case MSR_PAT:
-        val = env->pat;
-        break;
-    case MSR_IA32_PKRS:
-        val = env->pkrs;
-        break;
-    case MSR_VM_HSAVE_PA:
-        val = env->vm_hsave;
-        break;
-    case MSR_IA32_PERF_STATUS:
-        /* tsc_increment_by_tick */
-        val = 1000ULL;
-        /* CPU multiplier */
-        val |= (((uint64_t)4ULL) << 40);
-        break;
-#ifdef TARGET_X86_64
-    case MSR_LSTAR:
-        val = env->lstar;
-        break;
-    case MSR_CSTAR:
-        val = env->cstar;
-        break;
-    case MSR_FMASK:
-        val = env->fmask;
-        break;
-    case MSR_FSBASE:
-        val = env->segs[R_FS].base;
-        break;
-    case MSR_GSBASE:
-        val = env->segs[R_GS].base;
-        break;
-    case MSR_KERNELGSBASE:
-        val = env->kernelgsbase;
-        break;
-    case MSR_TSC_AUX:
-        val = env->tsc_aux;
-        break;
-#endif
-    case MSR_SMI_COUNT:
-        val = env->msr_smi_count;
-        break;
-    case MSR_MTRRphysBase(0):
-    case MSR_MTRRphysBase(1):
-    case MSR_MTRRphysBase(2):
-    case MSR_MTRRphysBase(3):
-    case MSR_MTRRphysBase(4):
-    case MSR_MTRRphysBase(5):
-    case MSR_MTRRphysBase(6):
-    case MSR_MTRRphysBase(7):
-        val = env->mtrr_var[((uint32_t)env->regs[R_ECX] -
-                             MSR_MTRRphysBase(0)) / 2].base;
-        break;
-    case MSR_MTRRphysMask(0):
-    case MSR_MTRRphysMask(1):
-    case MSR_MTRRphysMask(2):
-    case MSR_MTRRphysMask(3):
-    case MSR_MTRRphysMask(4):
-    case MSR_MTRRphysMask(5):
-    case MSR_MTRRphysMask(6):
-    case MSR_MTRRphysMask(7):
-        val = env->mtrr_var[((uint32_t)env->regs[R_ECX] -
-                             MSR_MTRRphysMask(0)) / 2].mask;
-        break;
-    case MSR_MTRRfix64K_00000:
-        val = env->mtrr_fixed[0];
-        break;
-    case MSR_MTRRfix16K_80000:
-    case MSR_MTRRfix16K_A0000:
-        val = env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
-                              MSR_MTRRfix16K_80000 + 1];
-        break;
-    case MSR_MTRRfix4K_C0000:
-    case MSR_MTRRfix4K_C8000:
-    case MSR_MTRRfix4K_D0000:
-    case MSR_MTRRfix4K_D8000:
-    case MSR_MTRRfix4K_E0000:
-    case MSR_MTRRfix4K_E8000:
-    case MSR_MTRRfix4K_F0000:
-    case MSR_MTRRfix4K_F8000:
-        val = env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
-                              MSR_MTRRfix4K_C0000 + 3];
-        break;
-    case MSR_MTRRdefType:
-        val = env->mtrr_deftype;
-        break;
-    case MSR_MTRRcap:
-        if (env->features[FEAT_1_EDX] & CPUID_MTRR) {
-            val = MSR_MTRRcap_VCNT | MSR_MTRRcap_FIXRANGE_SUPPORT |
-                MSR_MTRRcap_WC_SUPPORTED;
-        } else {
-            /* XXX: exception? */
-            val = 0;
-        }
-        break;
-    case MSR_MCG_CAP:
-        val = env->mcg_cap;
-        break;
-    case MSR_MCG_CTL:
-        if (env->mcg_cap & MCG_CTL_P) {
-            val = env->mcg_ctl;
-        } else {
-            val = 0;
-        }
-        break;
-    case MSR_MCG_STATUS:
-        val = env->mcg_status;
-        break;
-    case MSR_IA32_MISC_ENABLE:
-        val = env->msr_ia32_misc_enable;
-        break;
-    case MSR_IA32_BNDCFGS:
-        val = env->msr_bndcfgs;
-        break;
-     case MSR_IA32_UCODE_REV:
-        val = x86_cpu->ucode_rev;
-        break;
-    default:
-        if ((uint32_t)env->regs[R_ECX] >= MSR_MC0_CTL
-            && (uint32_t)env->regs[R_ECX] < MSR_MC0_CTL +
-            (4 * env->mcg_cap & 0xff)) {
-            uint32_t offset = (uint32_t)env->regs[R_ECX] - MSR_MC0_CTL;
-            val = env->mce_banks[offset];
-            break;
-        }
-        /* XXX: exception? */
-        val = 0;
-        break;
-    }
-    env->regs[R_EAX] = (uint32_t)(val);
-    env->regs[R_EDX] = (uint32_t)(val >> 32);
-}
-#endif
-
 static void do_pause(X86CPU *cpu)
 {
     CPUState *cs = CPU(cpu);
diff --git a/target/i386/tcg/softmmu/misc_helper.c b/target/i386/tcg/softmmu/misc_helper.c
new file mode 100644
index 0000000000..0a1b3be0c4
--- /dev/null
+++ b/target/i386/tcg/softmmu/misc_helper.c
@@ -0,0 +1,438 @@
+/*
+ *  x86 misc helpers - softmmu-only
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/main-loop.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+#include "exec/cpu_ldst.h"
+#include "exec/address-spaces.h"
+#include "tcg/helper-tcg.h"
+
+void helper_outb(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    address_space_stb(&address_space_io, port, data,
+                      cpu_get_mem_attrs(env), NULL);
+}
+
+target_ulong helper_inb(CPUX86State *env, uint32_t port)
+{
+    return address_space_ldub(&address_space_io, port,
+                              cpu_get_mem_attrs(env), NULL);
+}
+
+void helper_outw(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    address_space_stw(&address_space_io, port, data,
+                      cpu_get_mem_attrs(env), NULL);
+}
+
+target_ulong helper_inw(CPUX86State *env, uint32_t port)
+{
+    return address_space_lduw(&address_space_io, port,
+                              cpu_get_mem_attrs(env), NULL);
+}
+
+void helper_outl(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    address_space_stl(&address_space_io, port, data,
+                      cpu_get_mem_attrs(env), NULL);
+}
+
+target_ulong helper_inl(CPUX86State *env, uint32_t port)
+{
+    return address_space_ldl(&address_space_io, port,
+                             cpu_get_mem_attrs(env), NULL);
+}
+
+target_ulong helper_read_crN(CPUX86State *env, int reg)
+{
+    target_ulong val;
+
+    cpu_svm_check_intercept_param(env, SVM_EXIT_READ_CR0 + reg, 0, GETPC());
+    switch (reg) {
+    default:
+        val = env->cr[reg];
+        break;
+    case 8:
+        if (!(env->hflags2 & HF2_VINTR_MASK)) {
+            val = cpu_get_apic_tpr(env_archcpu(env)->apic_state);
+        } else {
+            val = env->v_tpr;
+        }
+        break;
+    }
+    return val;
+}
+
+void helper_write_crN(CPUX86State *env, int reg, target_ulong t0)
+{
+    cpu_svm_check_intercept_param(env, SVM_EXIT_WRITE_CR0 + reg, 0, GETPC());
+    switch (reg) {
+    case 0:
+        cpu_x86_update_cr0(env, t0);
+        break;
+    case 3:
+        cpu_x86_update_cr3(env, t0);
+        break;
+    case 4:
+        cpu_x86_update_cr4(env, t0);
+        break;
+    case 8:
+        if (!(env->hflags2 & HF2_VINTR_MASK)) {
+            qemu_mutex_lock_iothread();
+            cpu_set_apic_tpr(env_archcpu(env)->apic_state, t0);
+            qemu_mutex_unlock_iothread();
+        }
+        env->v_tpr = t0 & 0x0f;
+        break;
+    default:
+        env->cr[reg] = t0;
+        break;
+    }
+}
+
+void helper_wrmsr(CPUX86State *env)
+{
+    uint64_t val;
+    CPUState *cs = env_cpu(env);
+
+    cpu_svm_check_intercept_param(env, SVM_EXIT_MSR, 1, GETPC());
+
+    val = ((uint32_t)env->regs[R_EAX]) |
+        ((uint64_t)((uint32_t)env->regs[R_EDX]) << 32);
+
+    switch ((uint32_t)env->regs[R_ECX]) {
+    case MSR_IA32_SYSENTER_CS:
+        env->sysenter_cs = val & 0xffff;
+        break;
+    case MSR_IA32_SYSENTER_ESP:
+        env->sysenter_esp = val;
+        break;
+    case MSR_IA32_SYSENTER_EIP:
+        env->sysenter_eip = val;
+        break;
+    case MSR_IA32_APICBASE:
+        cpu_set_apic_base(env_archcpu(env)->apic_state, val);
+        break;
+    case MSR_EFER:
+        {
+            uint64_t update_mask;
+
+            update_mask = 0;
+            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_SYSCALL) {
+                update_mask |= MSR_EFER_SCE;
+            }
+            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM) {
+                update_mask |= MSR_EFER_LME;
+            }
+            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_FFXSR) {
+                update_mask |= MSR_EFER_FFXSR;
+            }
+            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_NX) {
+                update_mask |= MSR_EFER_NXE;
+            }
+            if (env->features[FEAT_8000_0001_ECX] & CPUID_EXT3_SVM) {
+                update_mask |= MSR_EFER_SVME;
+            }
+            if (env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_FFXSR) {
+                update_mask |= MSR_EFER_FFXSR;
+            }
+            cpu_load_efer(env, (env->efer & ~update_mask) |
+                          (val & update_mask));
+        }
+        break;
+    case MSR_STAR:
+        env->star = val;
+        break;
+    case MSR_PAT:
+        env->pat = val;
+        break;
+    case MSR_IA32_PKRS:
+        if (val & 0xFFFFFFFF00000000ull) {
+            goto error;
+        }
+        env->pkrs = val;
+        tlb_flush(cs);
+        break;
+    case MSR_VM_HSAVE_PA:
+        env->vm_hsave = val;
+        break;
+#ifdef TARGET_X86_64
+    case MSR_LSTAR:
+        env->lstar = val;
+        break;
+    case MSR_CSTAR:
+        env->cstar = val;
+        break;
+    case MSR_FMASK:
+        env->fmask = val;
+        break;
+    case MSR_FSBASE:
+        env->segs[R_FS].base = val;
+        break;
+    case MSR_GSBASE:
+        env->segs[R_GS].base = val;
+        break;
+    case MSR_KERNELGSBASE:
+        env->kernelgsbase = val;
+        break;
+#endif
+    case MSR_MTRRphysBase(0):
+    case MSR_MTRRphysBase(1):
+    case MSR_MTRRphysBase(2):
+    case MSR_MTRRphysBase(3):
+    case MSR_MTRRphysBase(4):
+    case MSR_MTRRphysBase(5):
+    case MSR_MTRRphysBase(6):
+    case MSR_MTRRphysBase(7):
+        env->mtrr_var[((uint32_t)env->regs[R_ECX] -
+                       MSR_MTRRphysBase(0)) / 2].base = val;
+        break;
+    case MSR_MTRRphysMask(0):
+    case MSR_MTRRphysMask(1):
+    case MSR_MTRRphysMask(2):
+    case MSR_MTRRphysMask(3):
+    case MSR_MTRRphysMask(4):
+    case MSR_MTRRphysMask(5):
+    case MSR_MTRRphysMask(6):
+    case MSR_MTRRphysMask(7):
+        env->mtrr_var[((uint32_t)env->regs[R_ECX] -
+                       MSR_MTRRphysMask(0)) / 2].mask = val;
+        break;
+    case MSR_MTRRfix64K_00000:
+        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
+                        MSR_MTRRfix64K_00000] = val;
+        break;
+    case MSR_MTRRfix16K_80000:
+    case MSR_MTRRfix16K_A0000:
+        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
+                        MSR_MTRRfix16K_80000 + 1] = val;
+        break;
+    case MSR_MTRRfix4K_C0000:
+    case MSR_MTRRfix4K_C8000:
+    case MSR_MTRRfix4K_D0000:
+    case MSR_MTRRfix4K_D8000:
+    case MSR_MTRRfix4K_E0000:
+    case MSR_MTRRfix4K_E8000:
+    case MSR_MTRRfix4K_F0000:
+    case MSR_MTRRfix4K_F8000:
+        env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
+                        MSR_MTRRfix4K_C0000 + 3] = val;
+        break;
+    case MSR_MTRRdefType:
+        env->mtrr_deftype = val;
+        break;
+    case MSR_MCG_STATUS:
+        env->mcg_status = val;
+        break;
+    case MSR_MCG_CTL:
+        if ((env->mcg_cap & MCG_CTL_P)
+            && (val == 0 || val == ~(uint64_t)0)) {
+            env->mcg_ctl = val;
+        }
+        break;
+    case MSR_TSC_AUX:
+        env->tsc_aux = val;
+        break;
+    case MSR_IA32_MISC_ENABLE:
+        env->msr_ia32_misc_enable = val;
+        break;
+    case MSR_IA32_BNDCFGS:
+        /* FIXME: #GP if reserved bits are set.  */
+        /* FIXME: Extend highest implemented bit of linear address.  */
+        env->msr_bndcfgs = val;
+        cpu_sync_bndcs_hflags(env);
+        break;
+    default:
+        if ((uint32_t)env->regs[R_ECX] >= MSR_MC0_CTL
+            && (uint32_t)env->regs[R_ECX] < MSR_MC0_CTL +
+            (4 * env->mcg_cap & 0xff)) {
+            uint32_t offset = (uint32_t)env->regs[R_ECX] - MSR_MC0_CTL;
+            if ((offset & 0x3) != 0
+                || (val == 0 || val == ~(uint64_t)0)) {
+                env->mce_banks[offset] = val;
+            }
+            break;
+        }
+        /* XXX: exception? */
+        break;
+    }
+    return;
+error:
+    raise_exception_err_ra(env, EXCP0D_GPF, 0, GETPC());
+}
+
+void helper_rdmsr(CPUX86State *env)
+{
+    X86CPU *x86_cpu = env_archcpu(env);
+    uint64_t val;
+
+    cpu_svm_check_intercept_param(env, SVM_EXIT_MSR, 0, GETPC());
+
+    switch ((uint32_t)env->regs[R_ECX]) {
+    case MSR_IA32_SYSENTER_CS:
+        val = env->sysenter_cs;
+        break;
+    case MSR_IA32_SYSENTER_ESP:
+        val = env->sysenter_esp;
+        break;
+    case MSR_IA32_SYSENTER_EIP:
+        val = env->sysenter_eip;
+        break;
+    case MSR_IA32_APICBASE:
+        val = cpu_get_apic_base(env_archcpu(env)->apic_state);
+        break;
+    case MSR_EFER:
+        val = env->efer;
+        break;
+    case MSR_STAR:
+        val = env->star;
+        break;
+    case MSR_PAT:
+        val = env->pat;
+        break;
+    case MSR_IA32_PKRS:
+        val = env->pkrs;
+        break;
+    case MSR_VM_HSAVE_PA:
+        val = env->vm_hsave;
+        break;
+    case MSR_IA32_PERF_STATUS:
+        /* tsc_increment_by_tick */
+        val = 1000ULL;
+        /* CPU multiplier */
+        val |= (((uint64_t)4ULL) << 40);
+        break;
+#ifdef TARGET_X86_64
+    case MSR_LSTAR:
+        val = env->lstar;
+        break;
+    case MSR_CSTAR:
+        val = env->cstar;
+        break;
+    case MSR_FMASK:
+        val = env->fmask;
+        break;
+    case MSR_FSBASE:
+        val = env->segs[R_FS].base;
+        break;
+    case MSR_GSBASE:
+        val = env->segs[R_GS].base;
+        break;
+    case MSR_KERNELGSBASE:
+        val = env->kernelgsbase;
+        break;
+    case MSR_TSC_AUX:
+        val = env->tsc_aux;
+        break;
+#endif
+    case MSR_SMI_COUNT:
+        val = env->msr_smi_count;
+        break;
+    case MSR_MTRRphysBase(0):
+    case MSR_MTRRphysBase(1):
+    case MSR_MTRRphysBase(2):
+    case MSR_MTRRphysBase(3):
+    case MSR_MTRRphysBase(4):
+    case MSR_MTRRphysBase(5):
+    case MSR_MTRRphysBase(6):
+    case MSR_MTRRphysBase(7):
+        val = env->mtrr_var[((uint32_t)env->regs[R_ECX] -
+                             MSR_MTRRphysBase(0)) / 2].base;
+        break;
+    case MSR_MTRRphysMask(0):
+    case MSR_MTRRphysMask(1):
+    case MSR_MTRRphysMask(2):
+    case MSR_MTRRphysMask(3):
+    case MSR_MTRRphysMask(4):
+    case MSR_MTRRphysMask(5):
+    case MSR_MTRRphysMask(6):
+    case MSR_MTRRphysMask(7):
+        val = env->mtrr_var[((uint32_t)env->regs[R_ECX] -
+                             MSR_MTRRphysMask(0)) / 2].mask;
+        break;
+    case MSR_MTRRfix64K_00000:
+        val = env->mtrr_fixed[0];
+        break;
+    case MSR_MTRRfix16K_80000:
+    case MSR_MTRRfix16K_A0000:
+        val = env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
+                              MSR_MTRRfix16K_80000 + 1];
+        break;
+    case MSR_MTRRfix4K_C0000:
+    case MSR_MTRRfix4K_C8000:
+    case MSR_MTRRfix4K_D0000:
+    case MSR_MTRRfix4K_D8000:
+    case MSR_MTRRfix4K_E0000:
+    case MSR_MTRRfix4K_E8000:
+    case MSR_MTRRfix4K_F0000:
+    case MSR_MTRRfix4K_F8000:
+        val = env->mtrr_fixed[(uint32_t)env->regs[R_ECX] -
+                              MSR_MTRRfix4K_C0000 + 3];
+        break;
+    case MSR_MTRRdefType:
+        val = env->mtrr_deftype;
+        break;
+    case MSR_MTRRcap:
+        if (env->features[FEAT_1_EDX] & CPUID_MTRR) {
+            val = MSR_MTRRcap_VCNT | MSR_MTRRcap_FIXRANGE_SUPPORT |
+                MSR_MTRRcap_WC_SUPPORTED;
+        } else {
+            /* XXX: exception? */
+            val = 0;
+        }
+        break;
+    case MSR_MCG_CAP:
+        val = env->mcg_cap;
+        break;
+    case MSR_MCG_CTL:
+        if (env->mcg_cap & MCG_CTL_P) {
+            val = env->mcg_ctl;
+        } else {
+            val = 0;
+        }
+        break;
+    case MSR_MCG_STATUS:
+        val = env->mcg_status;
+        break;
+    case MSR_IA32_MISC_ENABLE:
+        val = env->msr_ia32_misc_enable;
+        break;
+    case MSR_IA32_BNDCFGS:
+        val = env->msr_bndcfgs;
+        break;
+     case MSR_IA32_UCODE_REV:
+        val = x86_cpu->ucode_rev;
+        break;
+    default:
+        if ((uint32_t)env->regs[R_ECX] >= MSR_MC0_CTL
+            && (uint32_t)env->regs[R_ECX] < MSR_MC0_CTL +
+            (4 * env->mcg_cap & 0xff)) {
+            uint32_t offset = (uint32_t)env->regs[R_ECX] - MSR_MC0_CTL;
+            val = env->mce_banks[offset];
+            break;
+        }
+        /* XXX: exception? */
+        val = 0;
+        break;
+    }
+    env->regs[R_EAX] = (uint32_t)(val);
+    env->regs[R_EDX] = (uint32_t)(val >> 32);
+}
diff --git a/target/i386/tcg/user/misc_helper.c b/target/i386/tcg/user/misc_helper.c
new file mode 100644
index 0000000000..f5851c0389
--- /dev/null
+++ b/target/i386/tcg/user/misc_helper.c
@@ -0,0 +1,72 @@
+/*
+ *  x86 misc helpers
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+
+void helper_outb(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    fprintf(stderr, "outb: port=0x%04x, data=%02x\n", port, data);
+}
+
+target_ulong helper_inb(CPUX86State *env, uint32_t port)
+{
+    fprintf(stderr, "inb: port=0x%04x\n", port);
+    return 0;
+}
+
+void helper_outw(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    fprintf(stderr, "outw: port=0x%04x, data=%04x\n", port, data);
+}
+
+target_ulong helper_inw(CPUX86State *env, uint32_t port)
+{
+    fprintf(stderr, "inw: port=0x%04x\n", port);
+    return 0;
+}
+
+void helper_outl(CPUX86State *env, uint32_t port, uint32_t data)
+{
+    fprintf(stderr, "outl: port=0x%04x, data=%08x\n", port, data);
+}
+
+target_ulong helper_inl(CPUX86State *env, uint32_t port)
+{
+    fprintf(stderr, "inl: port=0x%04x\n", port);
+    return 0;
+}
+
+target_ulong helper_read_crN(CPUX86State *env, int reg)
+{
+    return 0;
+}
+
+void helper_write_crN(CPUX86State *env, int reg, target_ulong t0)
+{
+}
+
+void helper_wrmsr(CPUX86State *env)
+{
+}
+
+void helper_rdmsr(CPUX86State *env)
+{
+}
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index 1580950141..b2aaab6eef 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -3,4 +3,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'smm_helper.c',
   'excp_helper.c',
   'bpt_helper.c',
+  'misc_helper.c',
 ))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
index e0ef0f02e2..fb8cdc13ef 100644
--- a/target/i386/tcg/user/meson.build
+++ b/target/i386/tcg/user/meson.build
@@ -1,3 +1,4 @@
 i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
   'excp_helper.c',
+  'misc_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 12/15] i386: separate fpu_helper into user and softmmu parts
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (10 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 11/15] i386: split misc helper into user and softmmu parts Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-15 10:32   ` Alex Bennée
  2021-02-12 12:36 ` [RFC v18 13/15] i386: slit svm_helper into softmmu and stub-only user Claudio Fontana
                   ` (4 subsequent siblings)
  16 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/cpu.h                    |  3 ++
 target/i386/tcg/fpu_helper.c         | 65 +---------------------------
 target/i386/tcg/softmmu/fpu_helper.c | 58 +++++++++++++++++++++++++
 target/i386/tcg/user/fpu_helper.c    | 42 ++++++++++++++++++
 target/i386/tcg/softmmu/meson.build  |  1 +
 target/i386/tcg/user/meson.build     |  1 +
 6 files changed, 107 insertions(+), 63 deletions(-)
 create mode 100644 target/i386/tcg/softmmu/fpu_helper.c
 create mode 100644 target/i386/tcg/user/fpu_helper.c

diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index a4f9bbef55..afcf34e40b 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1811,7 +1811,10 @@ int cpu_x86_support_mca_broadcast(CPUX86State *env);
 int cpu_get_pic_interrupt(CPUX86State *s);
 /* MSDOS compatibility mode FPU exception support */
 void x86_register_ferr_irq(qemu_irq irq);
+bool fpu_check_raise_ferr_irq(CPUX86State *s);
 void cpu_set_ignne(void);
+void cpu_clear_ignne(void);
+
 /* mpx_helper.c */
 void cpu_sync_bndcs_hflags(CPUX86State *env);
 
diff --git a/target/i386/tcg/fpu_helper.c b/target/i386/tcg/fpu_helper.c
index 60ed93520a..257d121746 100644
--- a/target/i386/tcg/fpu_helper.c
+++ b/target/i386/tcg/fpu_helper.c
@@ -21,17 +21,10 @@
 #include <math.h>
 #include "cpu.h"
 #include "exec/helper-proto.h"
-#include "qemu/host-utils.h"
-#include "exec/exec-all.h"
-#include "exec/cpu_ldst.h"
 #include "fpu/softfloat.h"
 #include "fpu/softfloat-macros.h"
 #include "helper-tcg.h"
 
-#ifdef CONFIG_SOFTMMU
-#include "hw/irq.h"
-#endif
-
 /* float macros */
 #define FT0    (env->ft0)
 #define ST0    (env->fpregs[env->fpstt].d)
@@ -75,36 +68,6 @@
 #define floatx80_ln2_d make_floatx80(0x3ffe, 0xb17217f7d1cf79abLL)
 #define floatx80_pi_d make_floatx80(0x4000, 0xc90fdaa22168c234LL)
 
-#if !defined(CONFIG_USER_ONLY)
-static qemu_irq ferr_irq;
-
-void x86_register_ferr_irq(qemu_irq irq)
-{
-    ferr_irq = irq;
-}
-
-static void cpu_clear_ignne(void)
-{
-    CPUX86State *env = &X86_CPU(first_cpu)->env;
-    env->hflags2 &= ~HF2_IGNNE_MASK;
-}
-
-void cpu_set_ignne(void)
-{
-    CPUX86State *env = &X86_CPU(first_cpu)->env;
-    env->hflags2 |= HF2_IGNNE_MASK;
-    /*
-     * We get here in response to a write to port F0h.  The chipset should
-     * deassert FP_IRQ and FERR# instead should stay signaled until FPSW_SE is
-     * cleared, because FERR# and FP_IRQ are two separate pins on real
-     * hardware.  However, we don't model FERR# as a qemu_irq, so we just
-     * do directly what the chipset would do, i.e. deassert FP_IRQ.
-     */
-    qemu_irq_lower(ferr_irq);
-}
-#endif
-
-
 static inline void fpush(CPUX86State *env)
 {
     env->fpstt = (env->fpstt - 1) & 7;
@@ -203,8 +166,8 @@ static void fpu_raise_exception(CPUX86State *env, uintptr_t retaddr)
         raise_exception_ra(env, EXCP10_COPR, retaddr);
     }
 #if !defined(CONFIG_USER_ONLY)
-    else if (ferr_irq && !(env->hflags2 & HF2_IGNNE_MASK)) {
-        qemu_irq_raise(ferr_irq);
+    else {
+        (void)fpu_check_raise_ferr_irq(env);
     }
 #endif
 }
@@ -2501,18 +2464,6 @@ void helper_frstor(CPUX86State *env, target_ulong ptr, int data32)
     }
 }
 
-#if defined(CONFIG_USER_ONLY)
-void cpu_x86_fsave(CPUX86State *env, target_ulong ptr, int data32)
-{
-    helper_fsave(env, ptr, data32);
-}
-
-void cpu_x86_frstor(CPUX86State *env, target_ulong ptr, int data32)
-{
-    helper_frstor(env, ptr, data32);
-}
-#endif
-
 #define XO(X)  offsetof(X86XSaveArea, X)
 
 static void do_xsave_fpu(CPUX86State *env, target_ulong ptr, uintptr_t ra)
@@ -2780,18 +2731,6 @@ void helper_fxrstor(CPUX86State *env, target_ulong ptr)
     }
 }
 
-#if defined(CONFIG_USER_ONLY)
-void cpu_x86_fxsave(CPUX86State *env, target_ulong ptr)
-{
-    helper_fxsave(env, ptr);
-}
-
-void cpu_x86_fxrstor(CPUX86State *env, target_ulong ptr)
-{
-    helper_fxrstor(env, ptr);
-}
-#endif
-
 void helper_xrstor(CPUX86State *env, target_ulong ptr, uint64_t rfbm)
 {
     uintptr_t ra = GETPC();
diff --git a/target/i386/tcg/softmmu/fpu_helper.c b/target/i386/tcg/softmmu/fpu_helper.c
new file mode 100644
index 0000000000..eea9304973
--- /dev/null
+++ b/target/i386/tcg/softmmu/fpu_helper.c
@@ -0,0 +1,58 @@
+/*
+ *  x86 FPU, MMX/3DNow!/SSE/SSE2/SSE3/SSSE3/SSE4/PNI helpers (softmmu code)
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "hw/irq.h"
+
+static qemu_irq ferr_irq;
+
+void x86_register_ferr_irq(qemu_irq irq)
+{
+    ferr_irq = irq;
+}
+
+bool fpu_check_raise_ferr_irq(CPUX86State *env)
+{
+    if (ferr_irq && !(env->hflags2 & HF2_IGNNE_MASK)) {
+        qemu_irq_raise(ferr_irq);
+        return true;
+    }
+    return false;
+}
+
+void cpu_clear_ignne(void)
+{
+    CPUX86State *env = &X86_CPU(first_cpu)->env;
+    env->hflags2 &= ~HF2_IGNNE_MASK;
+}
+
+void cpu_set_ignne(void)
+{
+    CPUX86State *env = &X86_CPU(first_cpu)->env;
+    env->hflags2 |= HF2_IGNNE_MASK;
+    /*
+     * We get here in response to a write to port F0h.  The chipset should
+     * deassert FP_IRQ and FERR# instead should stay signaled until FPSW_SE is
+     * cleared, because FERR# and FP_IRQ are two separate pins on real
+     * hardware.  However, we don't model FERR# as a qemu_irq, so we just
+     * do directly what the chipset would do, i.e. deassert FP_IRQ.
+     */
+    qemu_irq_lower(ferr_irq);
+}
diff --git a/target/i386/tcg/user/fpu_helper.c b/target/i386/tcg/user/fpu_helper.c
new file mode 100644
index 0000000000..bb6f21fed8
--- /dev/null
+++ b/target/i386/tcg/user/fpu_helper.c
@@ -0,0 +1,42 @@
+/*
+ *  x86 FPU, MMX/3DNow!/SSE/SSE2/SSE3/SSSE3/SSE4/PNI helpers (user-mode)
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+
+void cpu_x86_fsave(CPUX86State *env, target_ulong ptr, int data32)
+{
+    helper_fsave(env, ptr, data32);
+}
+
+void cpu_x86_frstor(CPUX86State *env, target_ulong ptr, int data32)
+{
+    helper_frstor(env, ptr, data32);
+}
+
+void cpu_x86_fxsave(CPUX86State *env, target_ulong ptr)
+{
+    helper_fxsave(env, ptr);
+}
+
+void cpu_x86_fxrstor(CPUX86State *env, target_ulong ptr)
+{
+    helper_fxrstor(env, ptr);
+}
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index b2aaab6eef..f84519a213 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -4,4 +4,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'excp_helper.c',
   'bpt_helper.c',
   'misc_helper.c',
+  'fpu_helper.c',
 ))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
index fb8cdc13ef..30eec3f5a4 100644
--- a/target/i386/tcg/user/meson.build
+++ b/target/i386/tcg/user/meson.build
@@ -1,4 +1,5 @@
 i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
   'excp_helper.c',
   'misc_helper.c',
+  'fpu_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 13/15] i386: slit svm_helper into softmmu and stub-only user
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (11 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 12/15] i386: separate fpu_helper " Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 14/15] i386: split seg_helper into user-only and softmmu parts Claudio Fontana
                   ` (3 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/tcg/{ => softmmu}/svm_helper.c | 62 +-----------------
 target/i386/tcg/user/svm_stubs.c           | 76 ++++++++++++++++++++++
 target/i386/tcg/meson.build                |  1 -
 target/i386/tcg/softmmu/meson.build        |  1 +
 target/i386/tcg/user/meson.build           |  1 +
 5 files changed, 80 insertions(+), 61 deletions(-)
 rename target/i386/tcg/{ => softmmu}/svm_helper.c (96%)
 create mode 100644 target/i386/tcg/user/svm_stubs.c

diff --git a/target/i386/tcg/svm_helper.c b/target/i386/tcg/softmmu/svm_helper.c
similarity index 96%
rename from target/i386/tcg/svm_helper.c
rename to target/i386/tcg/softmmu/svm_helper.c
index 097bb9b83d..a2c9819330 100644
--- a/target/i386/tcg/svm_helper.c
+++ b/target/i386/tcg/softmmu/svm_helper.c
@@ -1,5 +1,5 @@
 /*
- *  x86 SVM helpers
+ *  x86 SVM helpers (softmmu-only)
  *
  *  Copyright (c) 2003 Fabrice Bellard
  *
@@ -22,66 +22,10 @@
 #include "exec/helper-proto.h"
 #include "exec/exec-all.h"
 #include "exec/cpu_ldst.h"
-#include "helper-tcg.h"
+#include "tcg/helper-tcg.h"
 
 /* Secure Virtual Machine helpers */
 
-#if defined(CONFIG_USER_ONLY)
-
-void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
-{
-}
-
-void helper_vmmcall(CPUX86State *env)
-{
-}
-
-void helper_vmload(CPUX86State *env, int aflag)
-{
-}
-
-void helper_vmsave(CPUX86State *env, int aflag)
-{
-}
-
-void helper_stgi(CPUX86State *env)
-{
-}
-
-void helper_clgi(CPUX86State *env)
-{
-}
-
-void helper_skinit(CPUX86State *env)
-{
-}
-
-void helper_invlpga(CPUX86State *env, int aflag)
-{
-}
-
-void cpu_vmexit(CPUX86State *nenv, uint32_t exit_code, uint64_t exit_info_1,
-                uintptr_t retaddr)
-{
-    assert(0);
-}
-
-void helper_svm_check_intercept_param(CPUX86State *env, uint32_t type,
-                                      uint64_t param)
-{
-}
-
-void cpu_svm_check_intercept_param(CPUX86State *env, uint32_t type,
-                                   uint64_t param, uintptr_t retaddr)
-{
-}
-
-void helper_svm_check_io(CPUX86State *env, uint32_t port, uint32_t param,
-                         uint32_t next_eip_addend)
-{
-}
-#else
-
 static inline void svm_save_seg(CPUX86State *env, hwaddr addr,
                                 const SegmentCache *sc)
 {
@@ -797,5 +741,3 @@ void do_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1)
        host's code segment or non-canonical (in the case of long mode), a
        #GP fault is delivered inside the host. */
 }
-
-#endif
diff --git a/target/i386/tcg/user/svm_stubs.c b/target/i386/tcg/user/svm_stubs.c
new file mode 100644
index 0000000000..97528b56ad
--- /dev/null
+++ b/target/i386/tcg/user/svm_stubs.c
@@ -0,0 +1,76 @@
+/*
+ *  x86 SVM helpers (user-mode)
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+#include "tcg/helper-tcg.h"
+
+void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
+{
+}
+
+void helper_vmmcall(CPUX86State *env)
+{
+}
+
+void helper_vmload(CPUX86State *env, int aflag)
+{
+}
+
+void helper_vmsave(CPUX86State *env, int aflag)
+{
+}
+
+void helper_stgi(CPUX86State *env)
+{
+}
+
+void helper_clgi(CPUX86State *env)
+{
+}
+
+void helper_skinit(CPUX86State *env)
+{
+}
+
+void helper_invlpga(CPUX86State *env, int aflag)
+{
+}
+
+void cpu_vmexit(CPUX86State *nenv, uint32_t exit_code, uint64_t exit_info_1,
+                uintptr_t retaddr)
+{
+    assert(0);
+}
+
+void helper_svm_check_intercept_param(CPUX86State *env, uint32_t type,
+                                      uint64_t param)
+{
+}
+
+void cpu_svm_check_intercept_param(CPUX86State *env, uint32_t type,
+                                   uint64_t param, uintptr_t retaddr)
+{
+}
+
+void helper_svm_check_io(CPUX86State *env, uint32_t port, uint32_t param,
+                         uint32_t next_eip_addend)
+{
+}
diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
index ec5daa1edc..6fbac2f240 100644
--- a/target/i386/tcg/meson.build
+++ b/target/i386/tcg/meson.build
@@ -8,7 +8,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
   'misc_helper.c',
   'mpx_helper.c',
   'seg_helper.c',
-  'svm_helper.c',
   'tcg-cpu.c',
   'translate.c'), if_false: files('tcg-stub.c'))
 
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index f84519a213..126528d0c9 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -5,4 +5,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'bpt_helper.c',
   'misc_helper.c',
   'fpu_helper.c',
+  'svm_helper.c',
 ))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
index 30eec3f5a4..c540ca2174 100644
--- a/target/i386/tcg/user/meson.build
+++ b/target/i386/tcg/user/meson.build
@@ -2,4 +2,5 @@ i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
   'excp_helper.c',
   'misc_helper.c',
   'fpu_helper.c',
+  'svm_stubs.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 14/15] i386: split seg_helper into user-only and softmmu parts
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (12 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 13/15] i386: slit svm_helper into softmmu and stub-only user Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:36 ` [RFC v18 15/15] i386: split off softmmu part of cpu.c Claudio Fontana
                   ` (2 subsequent siblings)
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/tcg/helper-tcg.h         |   5 +
 target/i386/tcg/seg_helper.h         |  66 ++++++++
 target/i386/tcg/seg_helper.c         | 233 +--------------------------
 target/i386/tcg/softmmu/seg_helper.c | 125 ++++++++++++++
 target/i386/tcg/user/seg_helper.c    | 109 +++++++++++++
 target/i386/tcg/softmmu/meson.build  |   1 +
 target/i386/tcg/user/meson.build     |   1 +
 7 files changed, 311 insertions(+), 229 deletions(-)
 create mode 100644 target/i386/tcg/seg_helper.h
 create mode 100644 target/i386/tcg/softmmu/seg_helper.c
 create mode 100644 target/i386/tcg/user/seg_helper.c

diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h
index b420b3356d..30eacdbbc9 100644
--- a/target/i386/tcg/helper-tcg.h
+++ b/target/i386/tcg/helper-tcg.h
@@ -88,6 +88,11 @@ void do_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1);
 
 /* seg_helper.c */
 void do_interrupt_x86_hardirq(CPUX86State *env, int intno, int is_hw);
+void do_interrupt_all(X86CPU *cpu, int intno, int is_int,
+                      int error_code, target_ulong next_eip, int is_hw);
+void handle_even_inj(CPUX86State *env, int intno, int is_int,
+                     int error_code, int is_hw, int rm);
+int exception_has_error_code(int intno);
 
 /* smm_helper.c */
 void do_smm_enter(X86CPU *cpu);
diff --git a/target/i386/tcg/seg_helper.h b/target/i386/tcg/seg_helper.h
new file mode 100644
index 0000000000..ebf1035277
--- /dev/null
+++ b/target/i386/tcg/seg_helper.h
@@ -0,0 +1,66 @@
+/*
+ *  x86 segmentation related helpers macros
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef SEG_HELPER_H
+#define SEG_HELPER_H
+
+//#define DEBUG_PCALL
+
+#ifdef DEBUG_PCALL
+# define LOG_PCALL(...) qemu_log_mask(CPU_LOG_PCALL, ## __VA_ARGS__)
+# define LOG_PCALL_STATE(cpu)                                  \
+    log_cpu_state_mask(CPU_LOG_PCALL, (cpu), CPU_DUMP_CCOP)
+#else
+# define LOG_PCALL(...) do { } while (0)
+# define LOG_PCALL_STATE(cpu) do { } while (0)
+#endif
+
+/*
+ * TODO: Convert callers to compute cpu_mmu_index_kernel once
+ * and use *_mmuidx_ra directly.
+ */
+#define cpu_ldub_kernel_ra(e, p, r) \
+    cpu_ldub_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+#define cpu_lduw_kernel_ra(e, p, r) \
+    cpu_lduw_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+#define cpu_ldl_kernel_ra(e, p, r) \
+    cpu_ldl_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+#define cpu_ldq_kernel_ra(e, p, r) \
+    cpu_ldq_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
+
+#define cpu_stb_kernel_ra(e, p, v, r) \
+    cpu_stb_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+#define cpu_stw_kernel_ra(e, p, v, r) \
+    cpu_stw_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+#define cpu_stl_kernel_ra(e, p, v, r) \
+    cpu_stl_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+#define cpu_stq_kernel_ra(e, p, v, r) \
+    cpu_stq_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
+
+#define cpu_ldub_kernel(e, p)    cpu_ldub_kernel_ra(e, p, 0)
+#define cpu_lduw_kernel(e, p)    cpu_lduw_kernel_ra(e, p, 0)
+#define cpu_ldl_kernel(e, p)     cpu_ldl_kernel_ra(e, p, 0)
+#define cpu_ldq_kernel(e, p)     cpu_ldq_kernel_ra(e, p, 0)
+
+#define cpu_stb_kernel(e, p, v)  cpu_stb_kernel_ra(e, p, v, 0)
+#define cpu_stw_kernel(e, p, v)  cpu_stw_kernel_ra(e, p, v, 0)
+#define cpu_stl_kernel(e, p, v)  cpu_stl_kernel_ra(e, p, v, 0)
+#define cpu_stq_kernel(e, p, v)  cpu_stq_kernel_ra(e, p, v, 0)
+
+#endif /* SEG_HELPER_H */
diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
index f0cb1bffe7..eb89383f4e 100644
--- a/target/i386/tcg/seg_helper.c
+++ b/target/i386/tcg/seg_helper.c
@@ -26,49 +26,7 @@
 #include "exec/cpu_ldst.h"
 #include "exec/log.h"
 #include "helper-tcg.h"
-
-//#define DEBUG_PCALL
-
-#ifdef DEBUG_PCALL
-# define LOG_PCALL(...) qemu_log_mask(CPU_LOG_PCALL, ## __VA_ARGS__)
-# define LOG_PCALL_STATE(cpu)                                  \
-    log_cpu_state_mask(CPU_LOG_PCALL, (cpu), CPU_DUMP_CCOP)
-#else
-# define LOG_PCALL(...) do { } while (0)
-# define LOG_PCALL_STATE(cpu) do { } while (0)
-#endif
-
-/*
- * TODO: Convert callers to compute cpu_mmu_index_kernel once
- * and use *_mmuidx_ra directly.
- */
-#define cpu_ldub_kernel_ra(e, p, r) \
-    cpu_ldub_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
-#define cpu_lduw_kernel_ra(e, p, r) \
-    cpu_lduw_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
-#define cpu_ldl_kernel_ra(e, p, r) \
-    cpu_ldl_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
-#define cpu_ldq_kernel_ra(e, p, r) \
-    cpu_ldq_mmuidx_ra(e, p, cpu_mmu_index_kernel(e), r)
-
-#define cpu_stb_kernel_ra(e, p, v, r) \
-    cpu_stb_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
-#define cpu_stw_kernel_ra(e, p, v, r) \
-    cpu_stw_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
-#define cpu_stl_kernel_ra(e, p, v, r) \
-    cpu_stl_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
-#define cpu_stq_kernel_ra(e, p, v, r) \
-    cpu_stq_mmuidx_ra(e, p, v, cpu_mmu_index_kernel(e), r)
-
-#define cpu_ldub_kernel(e, p)    cpu_ldub_kernel_ra(e, p, 0)
-#define cpu_lduw_kernel(e, p)    cpu_lduw_kernel_ra(e, p, 0)
-#define cpu_ldl_kernel(e, p)     cpu_ldl_kernel_ra(e, p, 0)
-#define cpu_ldq_kernel(e, p)     cpu_ldq_kernel_ra(e, p, 0)
-
-#define cpu_stb_kernel(e, p, v)  cpu_stb_kernel_ra(e, p, v, 0)
-#define cpu_stw_kernel(e, p, v)  cpu_stw_kernel_ra(e, p, v, 0)
-#define cpu_stl_kernel(e, p, v)  cpu_stl_kernel_ra(e, p, v, 0)
-#define cpu_stq_kernel(e, p, v)  cpu_stq_kernel_ra(e, p, v, 0)
+#include "seg_helper.h"
 
 /* return non zero if error */
 static inline int load_segment_ra(CPUX86State *env, uint32_t *e1_ptr,
@@ -531,7 +489,7 @@ static inline unsigned int get_sp_mask(unsigned int e2)
     }
 }
 
-static int exception_has_error_code(int intno)
+int exception_has_error_code(int intno)
 {
     switch (intno) {
     case 8:
@@ -976,72 +934,6 @@ static void do_interrupt64(CPUX86State *env, int intno, int is_int,
 }
 #endif
 
-#ifdef TARGET_X86_64
-#if defined(CONFIG_USER_ONLY)
-void helper_syscall(CPUX86State *env, int next_eip_addend)
-{
-    CPUState *cs = env_cpu(env);
-
-    cs->exception_index = EXCP_SYSCALL;
-    env->exception_is_int = 0;
-    env->exception_next_eip = env->eip + next_eip_addend;
-    cpu_loop_exit(cs);
-}
-#else
-void helper_syscall(CPUX86State *env, int next_eip_addend)
-{
-    int selector;
-
-    if (!(env->efer & MSR_EFER_SCE)) {
-        raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
-    }
-    selector = (env->star >> 32) & 0xffff;
-    if (env->hflags & HF_LMA_MASK) {
-        int code64;
-
-        env->regs[R_ECX] = env->eip + next_eip_addend;
-        env->regs[11] = cpu_compute_eflags(env) & ~RF_MASK;
-
-        code64 = env->hflags & HF_CS64_MASK;
-
-        env->eflags &= ~(env->fmask | RF_MASK);
-        cpu_load_eflags(env, env->eflags, 0);
-        cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
-                           0, 0xffffffff,
-                               DESC_G_MASK | DESC_P_MASK |
-                               DESC_S_MASK |
-                               DESC_CS_MASK | DESC_R_MASK | DESC_A_MASK |
-                               DESC_L_MASK);
-        cpu_x86_load_seg_cache(env, R_SS, (selector + 8) & 0xfffc,
-                               0, 0xffffffff,
-                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
-                               DESC_S_MASK |
-                               DESC_W_MASK | DESC_A_MASK);
-        if (code64) {
-            env->eip = env->lstar;
-        } else {
-            env->eip = env->cstar;
-        }
-    } else {
-        env->regs[R_ECX] = (uint32_t)(env->eip + next_eip_addend);
-
-        env->eflags &= ~(IF_MASK | RF_MASK | VM_MASK);
-        cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
-                           0, 0xffffffff,
-                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
-                               DESC_S_MASK |
-                               DESC_CS_MASK | DESC_R_MASK | DESC_A_MASK);
-        cpu_x86_load_seg_cache(env, R_SS, (selector + 8) & 0xfffc,
-                               0, 0xffffffff,
-                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
-                               DESC_S_MASK |
-                               DESC_W_MASK | DESC_A_MASK);
-        env->eip = (uint32_t)env->star;
-    }
-}
-#endif
-#endif
-
 #ifdef TARGET_X86_64
 void helper_sysret(CPUX86State *env, int dflag)
 {
@@ -1136,84 +1028,13 @@ static void do_interrupt_real(CPUX86State *env, int intno, int is_int,
     env->eflags &= ~(IF_MASK | TF_MASK | AC_MASK | RF_MASK);
 }
 
-#if defined(CONFIG_USER_ONLY)
-/* fake user mode interrupt. is_int is TRUE if coming from the int
- * instruction. next_eip is the env->eip value AFTER the interrupt
- * instruction. It is only relevant if is_int is TRUE or if intno
- * is EXCP_SYSCALL.
- */
-static void do_interrupt_user(CPUX86State *env, int intno, int is_int,
-                              int error_code, target_ulong next_eip)
-{
-    if (is_int) {
-        SegmentCache *dt;
-        target_ulong ptr;
-        int dpl, cpl, shift;
-        uint32_t e2;
-
-        dt = &env->idt;
-        if (env->hflags & HF_LMA_MASK) {
-            shift = 4;
-        } else {
-            shift = 3;
-        }
-        ptr = dt->base + (intno << shift);
-        e2 = cpu_ldl_kernel(env, ptr + 4);
-
-        dpl = (e2 >> DESC_DPL_SHIFT) & 3;
-        cpl = env->hflags & HF_CPL_MASK;
-        /* check privilege if software int */
-        if (dpl < cpl) {
-            raise_exception_err(env, EXCP0D_GPF, (intno << shift) + 2);
-        }
-    }
-
-    /* Since we emulate only user space, we cannot do more than
-       exiting the emulation with the suitable exception and error
-       code. So update EIP for INT 0x80 and EXCP_SYSCALL. */
-    if (is_int || intno == EXCP_SYSCALL) {
-        env->eip = next_eip;
-    }
-}
-
-#else
-
-static void handle_even_inj(CPUX86State *env, int intno, int is_int,
-                            int error_code, int is_hw, int rm)
-{
-    CPUState *cs = env_cpu(env);
-    uint32_t event_inj = x86_ldl_phys(cs, env->vm_vmcb + offsetof(struct vmcb,
-                                                          control.event_inj));
-
-    if (!(event_inj & SVM_EVTINJ_VALID)) {
-        int type;
-
-        if (is_int) {
-            type = SVM_EVTINJ_TYPE_SOFT;
-        } else {
-            type = SVM_EVTINJ_TYPE_EXEPT;
-        }
-        event_inj = intno | type | SVM_EVTINJ_VALID;
-        if (!rm && exception_has_error_code(intno)) {
-            event_inj |= SVM_EVTINJ_VALID_ERR;
-            x86_stl_phys(cs, env->vm_vmcb + offsetof(struct vmcb,
-                                             control.event_inj_err),
-                     error_code);
-        }
-        x86_stl_phys(cs,
-                 env->vm_vmcb + offsetof(struct vmcb, control.event_inj),
-                 event_inj);
-    }
-}
-#endif
-
 /*
  * Begin execution of an interruption. is_int is TRUE if coming from
  * the int instruction. next_eip is the env->eip value AFTER the interrupt
  * instruction. It is only relevant if is_int is TRUE.
  */
-static void do_interrupt_all(X86CPU *cpu, int intno, int is_int,
-                             int error_code, target_ulong next_eip, int is_hw)
+void do_interrupt_all(X86CPU *cpu, int intno, int is_int,
+                      int error_code, target_ulong next_eip, int is_hw)
 {
     CPUX86State *env = &cpu->env;
 
@@ -1289,36 +1110,6 @@ static void do_interrupt_all(X86CPU *cpu, int intno, int is_int,
 #endif
 }
 
-void x86_cpu_do_interrupt(CPUState *cs)
-{
-    X86CPU *cpu = X86_CPU(cs);
-    CPUX86State *env = &cpu->env;
-
-#if defined(CONFIG_USER_ONLY)
-    /* if user mode only, we simulate a fake exception
-       which will be handled outside the cpu execution
-       loop */
-    do_interrupt_user(env, cs->exception_index,
-                      env->exception_is_int,
-                      env->error_code,
-                      env->exception_next_eip);
-    /* successfully delivered */
-    env->old_exception = -1;
-#else
-    if (cs->exception_index >= EXCP_VMEXIT) {
-        assert(env->old_exception == -1);
-        do_vmexit(env, cs->exception_index - EXCP_VMEXIT, env->error_code);
-    } else {
-        do_interrupt_all(cpu, cs->exception_index,
-                         env->exception_is_int,
-                         env->error_code,
-                         env->exception_next_eip, 0);
-        /* successfully delivered */
-        env->old_exception = -1;
-    }
-#endif
-}
-
 void do_interrupt_x86_hardirq(CPUX86State *env, int intno, int is_hw)
 {
     do_interrupt_all(env_archcpu(env), intno, 0, 0, 0, is_hw);
@@ -2624,22 +2415,6 @@ void helper_verw(CPUX86State *env, target_ulong selector1)
     CC_SRC = eflags | CC_Z;
 }
 
-#if defined(CONFIG_USER_ONLY)
-void cpu_x86_load_seg(CPUX86State *env, X86Seg seg_reg, int selector)
-{
-    if (!(env->cr[0] & CR0_PE_MASK) || (env->eflags & VM_MASK)) {
-        int dpl = (env->eflags & VM_MASK) ? 3 : 0;
-        selector &= 0xffff;
-        cpu_x86_load_seg_cache(env, seg_reg, selector,
-                               (selector << 4), 0xffff,
-                               DESC_P_MASK | DESC_S_MASK | DESC_W_MASK |
-                               DESC_A_MASK | (dpl << DESC_DPL_SHIFT));
-    } else {
-        helper_load_seg(env, seg_reg, selector);
-    }
-}
-#endif
-
 /* check if Port I/O is allowed in TSS */
 static inline void check_io(CPUX86State *env, int addr, int size,
                             uintptr_t retaddr)
diff --git a/target/i386/tcg/softmmu/seg_helper.c b/target/i386/tcg/softmmu/seg_helper.c
new file mode 100644
index 0000000000..84dafc4544
--- /dev/null
+++ b/target/i386/tcg/softmmu/seg_helper.c
@@ -0,0 +1,125 @@
+/*
+ *  x86 segmentation related helpers: (softmmu-only code)
+ *  TSS, interrupts, system calls, jumps and call/task gates, descriptors
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+#include "exec/cpu_ldst.h"
+#include "tcg/helper-tcg.h"
+
+#ifdef TARGET_X86_64
+void helper_syscall(CPUX86State *env, int next_eip_addend)
+{
+    int selector;
+
+    if (!(env->efer & MSR_EFER_SCE)) {
+        raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
+    }
+    selector = (env->star >> 32) & 0xffff;
+    if (env->hflags & HF_LMA_MASK) {
+        int code64;
+
+        env->regs[R_ECX] = env->eip + next_eip_addend;
+        env->regs[11] = cpu_compute_eflags(env) & ~RF_MASK;
+
+        code64 = env->hflags & HF_CS64_MASK;
+
+        env->eflags &= ~(env->fmask | RF_MASK);
+        cpu_load_eflags(env, env->eflags, 0);
+        cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
+                           0, 0xffffffff,
+                               DESC_G_MASK | DESC_P_MASK |
+                               DESC_S_MASK |
+                               DESC_CS_MASK | DESC_R_MASK | DESC_A_MASK |
+                               DESC_L_MASK);
+        cpu_x86_load_seg_cache(env, R_SS, (selector + 8) & 0xfffc,
+                               0, 0xffffffff,
+                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
+                               DESC_S_MASK |
+                               DESC_W_MASK | DESC_A_MASK);
+        if (code64) {
+            env->eip = env->lstar;
+        } else {
+            env->eip = env->cstar;
+        }
+    } else {
+        env->regs[R_ECX] = (uint32_t)(env->eip + next_eip_addend);
+
+        env->eflags &= ~(IF_MASK | RF_MASK | VM_MASK);
+        cpu_x86_load_seg_cache(env, R_CS, selector & 0xfffc,
+                           0, 0xffffffff,
+                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
+                               DESC_S_MASK |
+                               DESC_CS_MASK | DESC_R_MASK | DESC_A_MASK);
+        cpu_x86_load_seg_cache(env, R_SS, (selector + 8) & 0xfffc,
+                               0, 0xffffffff,
+                               DESC_G_MASK | DESC_B_MASK | DESC_P_MASK |
+                               DESC_S_MASK |
+                               DESC_W_MASK | DESC_A_MASK);
+        env->eip = (uint32_t)env->star;
+    }
+}
+#endif /* TARGET_X86_64 */
+
+void handle_even_inj(CPUX86State *env, int intno, int is_int,
+                     int error_code, int is_hw, int rm)
+{
+    CPUState *cs = env_cpu(env);
+    uint32_t event_inj = x86_ldl_phys(cs, env->vm_vmcb + offsetof(struct vmcb,
+                                                          control.event_inj));
+
+    if (!(event_inj & SVM_EVTINJ_VALID)) {
+        int type;
+
+        if (is_int) {
+            type = SVM_EVTINJ_TYPE_SOFT;
+        } else {
+            type = SVM_EVTINJ_TYPE_EXEPT;
+        }
+        event_inj = intno | type | SVM_EVTINJ_VALID;
+        if (!rm && exception_has_error_code(intno)) {
+            event_inj |= SVM_EVTINJ_VALID_ERR;
+            x86_stl_phys(cs, env->vm_vmcb + offsetof(struct vmcb,
+                                             control.event_inj_err),
+                     error_code);
+        }
+        x86_stl_phys(cs,
+                 env->vm_vmcb + offsetof(struct vmcb, control.event_inj),
+                 event_inj);
+    }
+}
+
+void x86_cpu_do_interrupt(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    if (cs->exception_index >= EXCP_VMEXIT) {
+        assert(env->old_exception == -1);
+        do_vmexit(env, cs->exception_index - EXCP_VMEXIT, env->error_code);
+    } else {
+        do_interrupt_all(cpu, cs->exception_index,
+                         env->exception_is_int,
+                         env->error_code,
+                         env->exception_next_eip, 0);
+        /* successfully delivered */
+        env->old_exception = -1;
+    }
+}
diff --git a/target/i386/tcg/user/seg_helper.c b/target/i386/tcg/user/seg_helper.c
new file mode 100644
index 0000000000..67481b0aa8
--- /dev/null
+++ b/target/i386/tcg/user/seg_helper.c
@@ -0,0 +1,109 @@
+/*
+ *  x86 segmentation related helpers (user-mode code):
+ *  TSS, interrupts, system calls, jumps and call/task gates, descriptors
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+#include "exec/exec-all.h"
+#include "exec/cpu_ldst.h"
+#include "tcg/helper-tcg.h"
+#include "tcg/seg_helper.h"
+
+#ifdef TARGET_X86_64
+void helper_syscall(CPUX86State *env, int next_eip_addend)
+{
+    CPUState *cs = env_cpu(env);
+
+    cs->exception_index = EXCP_SYSCALL;
+    env->exception_is_int = 0;
+    env->exception_next_eip = env->eip + next_eip_addend;
+    cpu_loop_exit(cs);
+}
+#endif /* TARGET_X86_64 */
+
+/*
+ * fake user mode interrupt. is_int is TRUE if coming from the int
+ * instruction. next_eip is the env->eip value AFTER the interrupt
+ * instruction. It is only relevant if is_int is TRUE or if intno
+ * is EXCP_SYSCALL.
+ */
+static void do_interrupt_user(CPUX86State *env, int intno, int is_int,
+                              int error_code, target_ulong next_eip)
+{
+    if (is_int) {
+        SegmentCache *dt;
+        target_ulong ptr;
+        int dpl, cpl, shift;
+        uint32_t e2;
+
+        dt = &env->idt;
+        if (env->hflags & HF_LMA_MASK) {
+            shift = 4;
+        } else {
+            shift = 3;
+        }
+        ptr = dt->base + (intno << shift);
+        e2 = cpu_ldl_kernel(env, ptr + 4);
+
+        dpl = (e2 >> DESC_DPL_SHIFT) & 3;
+        cpl = env->hflags & HF_CPL_MASK;
+        /* check privilege if software int */
+        if (dpl < cpl) {
+            raise_exception_err(env, EXCP0D_GPF, (intno << shift) + 2);
+        }
+    }
+
+    /* Since we emulate only user space, we cannot do more than
+       exiting the emulation with the suitable exception and error
+       code. So update EIP for INT 0x80 and EXCP_SYSCALL. */
+    if (is_int || intno == EXCP_SYSCALL) {
+        env->eip = next_eip;
+    }
+}
+
+void x86_cpu_do_interrupt(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    /* if user mode only, we simulate a fake exception
+       which will be handled outside the cpu execution
+       loop */
+    do_interrupt_user(env, cs->exception_index,
+                      env->exception_is_int,
+                      env->error_code,
+                      env->exception_next_eip);
+    /* successfully delivered */
+    env->old_exception = -1;
+}
+
+void cpu_x86_load_seg(CPUX86State *env, X86Seg seg_reg, int selector)
+{
+    if (!(env->cr[0] & CR0_PE_MASK) || (env->eflags & VM_MASK)) {
+        int dpl = (env->eflags & VM_MASK) ? 3 : 0;
+        selector &= 0xffff;
+        cpu_x86_load_seg_cache(env, seg_reg, selector,
+                               (selector << 4), 0xffff,
+                               DESC_P_MASK | DESC_S_MASK | DESC_W_MASK |
+                               DESC_A_MASK | (dpl << DESC_DPL_SHIFT));
+    } else {
+        helper_load_seg(env, seg_reg, selector);
+    }
+}
diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
index 126528d0c9..2e444e766a 100644
--- a/target/i386/tcg/softmmu/meson.build
+++ b/target/i386/tcg/softmmu/meson.build
@@ -6,4 +6,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
   'misc_helper.c',
   'fpu_helper.c',
   'svm_helper.c',
+  'seg_helper.c',
 ))
diff --git a/target/i386/tcg/user/meson.build b/target/i386/tcg/user/meson.build
index c540ca2174..1f2801c45b 100644
--- a/target/i386/tcg/user/meson.build
+++ b/target/i386/tcg/user/meson.build
@@ -3,4 +3,5 @@ i386_user_ss.add(when: ['CONFIG_TCG', 'CONFIG_USER_ONLY'], if_true: files(
   'misc_helper.c',
   'fpu_helper.c',
   'svm_stubs.c',
+  'seg_helper.c',
 ))
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* [RFC v18 15/15] i386: split off softmmu part of cpu.c
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (13 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 14/15] i386: split seg_helper into user-only and softmmu parts Claudio Fontana
@ 2021-02-12 12:36 ` Claudio Fontana
  2021-02-12 12:57 ` [RFC v18 00/15] i386 cleanup PART 2 no-reply
  2021-02-15 11:37 ` Alex Bennée
  16 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-12 12:36 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, Claudio Fontana,
	qemu-devel

Signed-off-by: Claudio Fontana <cfontana@suse.de>
---
 target/i386/cpu-internal.h |  70 +++++++
 target/i386/cpu-softmmu.c  | 352 ++++++++++++++++++++++++++++++++++
 target/i386/cpu.c          | 383 +------------------------------------
 target/i386/meson.build    |   1 +
 4 files changed, 428 insertions(+), 378 deletions(-)
 create mode 100644 target/i386/cpu-internal.h
 create mode 100644 target/i386/cpu-softmmu.c

diff --git a/target/i386/cpu-internal.h b/target/i386/cpu-internal.h
new file mode 100644
index 0000000000..9ecadf593b
--- /dev/null
+++ b/target/i386/cpu-internal.h
@@ -0,0 +1,70 @@
+/*
+ * i386 CPU internal definitions to be shared between general and softmmu
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef I386_CPU_INTERNAL_H
+#define I386_CPU_INTERNAL_H
+
+typedef enum FeatureWordType {
+   CPUID_FEATURE_WORD,
+   MSR_FEATURE_WORD,
+} FeatureWordType;
+
+typedef struct FeatureWordInfo {
+    FeatureWordType type;
+    /* feature flags names are taken from "Intel Processor Identification and
+     * the CPUID Instruction" and AMD's "CPUID Specification".
+     * In cases of disagreement between feature naming conventions,
+     * aliases may be added.
+     */
+    const char *feat_names[64];
+    union {
+        /* If type==CPUID_FEATURE_WORD */
+        struct {
+            uint32_t eax;   /* Input EAX for CPUID */
+            bool needs_ecx; /* CPUID instruction uses ECX as input */
+            uint32_t ecx;   /* Input ECX value for CPUID */
+            int reg;        /* output register (R_* constant) */
+        } cpuid;
+        /* If type==MSR_FEATURE_WORD */
+        struct {
+            uint32_t index;
+        } msr;
+    };
+    uint64_t tcg_features; /* Feature flags supported by TCG */
+    uint64_t unmigratable_flags; /* Feature flags known to be unmigratable */
+    uint64_t migratable_flags; /* Feature flags known to be migratable */
+    /* Features that shouldn't be auto-enabled by "-cpu host" */
+    uint64_t no_autoenable_flags;
+} FeatureWordInfo;
+
+extern FeatureWordInfo feature_word_info[];
+
+void x86_cpu_expand_features(X86CPU *cpu, Error **errp);
+
+#ifndef CONFIG_USER_ONLY
+GuestPanicInformation *x86_cpu_get_crash_info(CPUState *cs);
+void x86_cpu_get_crash_info_qom(Object *obj, Visitor *v,
+                                const char *name, void *opaque, Error **errp);
+
+void x86_cpu_apic_create(X86CPU *cpu, Error **errp);
+void x86_cpu_apic_realize(X86CPU *cpu, Error **errp);
+void x86_cpu_machine_reset_cb(void *opaque);
+#endif /* CONFIG_USER_ONLY */
+
+#endif /* CPU_INTERNAL_H */
diff --git a/target/i386/cpu-softmmu.c b/target/i386/cpu-softmmu.c
new file mode 100644
index 0000000000..c824408e8e
--- /dev/null
+++ b/target/i386/cpu-softmmu.c
@@ -0,0 +1,352 @@
+/*
+ *  i386 CPUID helper functions
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "sysemu/xen.h"
+#include "sysemu/whpx.h"
+#include "kvm/kvm_i386.h"
+#include "qapi/error.h"
+#include "qapi/qapi-visit-run-state.h"
+#include "qapi/qmp/qdict.h"
+#include "qom/qom-qobject.h"
+#include "qapi/qapi-commands-machine-target.h"
+#include "hw/qdev-properties.h"
+
+#include "exec/address-spaces.h"
+#include "hw/i386/apic_internal.h"
+
+#include "cpu-internal.h"
+
+/* Return a QDict containing keys for all properties that can be included
+ * in static expansion of CPU models. All properties set by x86_cpu_load_model()
+ * must be included in the dictionary.
+ */
+static QDict *x86_cpu_static_props(void)
+{
+    FeatureWord w;
+    int i;
+    static const char *props[] = {
+        "min-level",
+        "min-xlevel",
+        "family",
+        "model",
+        "stepping",
+        "model-id",
+        "vendor",
+        "lmce",
+        NULL,
+    };
+    static QDict *d;
+
+    if (d) {
+        return d;
+    }
+
+    d = qdict_new();
+    for (i = 0; props[i]; i++) {
+        qdict_put_null(d, props[i]);
+    }
+
+    for (w = 0; w < FEATURE_WORDS; w++) {
+        FeatureWordInfo *fi = &feature_word_info[w];
+        int bit;
+        for (bit = 0; bit < 64; bit++) {
+            if (!fi->feat_names[bit]) {
+                continue;
+            }
+            qdict_put_null(d, fi->feat_names[bit]);
+        }
+    }
+
+    return d;
+}
+
+/* Add an entry to @props dict, with the value for property. */
+static void x86_cpu_expand_prop(X86CPU *cpu, QDict *props, const char *prop)
+{
+    QObject *value = object_property_get_qobject(OBJECT(cpu), prop,
+                                                 &error_abort);
+
+    qdict_put_obj(props, prop, value);
+}
+
+/* Convert CPU model data from X86CPU object to a property dictionary
+ * that can recreate exactly the same CPU model.
+ */
+static void x86_cpu_to_dict(X86CPU *cpu, QDict *props)
+{
+    QDict *sprops = x86_cpu_static_props();
+    const QDictEntry *e;
+
+    for (e = qdict_first(sprops); e; e = qdict_next(sprops, e)) {
+        const char *prop = qdict_entry_key(e);
+        x86_cpu_expand_prop(cpu, props, prop);
+    }
+}
+
+/* Convert CPU model data from X86CPU object to a property dictionary
+ * that can recreate exactly the same CPU model, including every
+ * writeable QOM property.
+ */
+static void x86_cpu_to_dict_full(X86CPU *cpu, QDict *props)
+{
+    ObjectPropertyIterator iter;
+    ObjectProperty *prop;
+
+    object_property_iter_init(&iter, OBJECT(cpu));
+    while ((prop = object_property_iter_next(&iter))) {
+        /* skip read-only or write-only properties */
+        if (!prop->get || !prop->set) {
+            continue;
+        }
+
+        /* "hotplugged" is the only property that is configurable
+         * on the command-line but will be set differently on CPUs
+         * created using "-cpu ... -smp ..." and by CPUs created
+         * on the fly by x86_cpu_from_model() for querying. Skip it.
+         */
+        if (!strcmp(prop->name, "hotplugged")) {
+            continue;
+        }
+        x86_cpu_expand_prop(cpu, props, prop->name);
+    }
+}
+
+static void object_apply_props(Object *obj, QDict *props, Error **errp)
+{
+    const QDictEntry *prop;
+
+    for (prop = qdict_first(props); prop; prop = qdict_next(props, prop)) {
+        if (!object_property_set_qobject(obj, qdict_entry_key(prop),
+                                         qdict_entry_value(prop), errp)) {
+            break;
+        }
+    }
+}
+
+/* Create X86CPU object according to model+props specification */
+static X86CPU *x86_cpu_from_model(const char *model, QDict *props, Error **errp)
+{
+    X86CPU *xc = NULL;
+    X86CPUClass *xcc;
+    Error *err = NULL;
+
+    xcc = X86_CPU_CLASS(cpu_class_by_name(TYPE_X86_CPU, model));
+    if (xcc == NULL) {
+        error_setg(&err, "CPU model '%s' not found", model);
+        goto out;
+    }
+
+    xc = X86_CPU(object_new_with_class(OBJECT_CLASS(xcc)));
+    if (props) {
+        object_apply_props(OBJECT(xc), props, &err);
+        if (err) {
+            goto out;
+        }
+    }
+
+    x86_cpu_expand_features(xc, &err);
+    if (err) {
+        goto out;
+    }
+
+out:
+    if (err) {
+        error_propagate(errp, err);
+        object_unref(OBJECT(xc));
+        xc = NULL;
+    }
+    return xc;
+}
+
+CpuModelExpansionInfo *
+qmp_query_cpu_model_expansion(CpuModelExpansionType type,
+                                                      CpuModelInfo *model,
+                                                      Error **errp)
+{
+    X86CPU *xc = NULL;
+    Error *err = NULL;
+    CpuModelExpansionInfo *ret = g_new0(CpuModelExpansionInfo, 1);
+    QDict *props = NULL;
+    const char *base_name;
+
+    xc = x86_cpu_from_model(model->name,
+                            model->has_props ?
+                                qobject_to(QDict, model->props) :
+                                NULL, &err);
+    if (err) {
+        goto out;
+    }
+
+    props = qdict_new();
+    ret->model = g_new0(CpuModelInfo, 1);
+    ret->model->props = QOBJECT(props);
+    ret->model->has_props = true;
+
+    switch (type) {
+    case CPU_MODEL_EXPANSION_TYPE_STATIC:
+        /* Static expansion will be based on "base" only */
+        base_name = "base";
+        x86_cpu_to_dict(xc, props);
+    break;
+    case CPU_MODEL_EXPANSION_TYPE_FULL:
+        /* As we don't return every single property, full expansion needs
+         * to keep the original model name+props, and add extra
+         * properties on top of that.
+         */
+        base_name = model->name;
+        x86_cpu_to_dict_full(xc, props);
+    break;
+    default:
+        error_setg(&err, "Unsupported expansion type");
+        goto out;
+    }
+
+    x86_cpu_to_dict(xc, props);
+
+    ret->model->name = g_strdup(base_name);
+
+out:
+    object_unref(OBJECT(xc));
+    if (err) {
+        error_propagate(errp, err);
+        qapi_free_CpuModelExpansionInfo(ret);
+        ret = NULL;
+    }
+    return ret;
+}
+
+void cpu_clear_apic_feature(CPUX86State *env)
+{
+    env->features[FEAT_1_EDX] &= ~CPUID_APIC;
+}
+
+bool cpu_is_bsp(X86CPU *cpu)
+{
+    return cpu_get_apic_base(cpu->apic_state) & MSR_IA32_APICBASE_BSP;
+}
+
+/* TODO: remove me, when reset over QOM tree is implemented */
+void x86_cpu_machine_reset_cb(void *opaque)
+{
+    X86CPU *cpu = opaque;
+    cpu_reset(CPU(cpu));
+}
+
+APICCommonClass *apic_get_class(void)
+{
+    const char *apic_type = "apic";
+
+    /* TODO: in-kernel irqchip for hvf */
+    if (kvm_apic_in_kernel()) {
+        apic_type = "kvm-apic";
+    } else if (xen_enabled()) {
+        apic_type = "xen-apic";
+    } else if (whpx_apic_in_platform()) {
+        apic_type = "whpx-apic";
+    }
+
+    return APIC_COMMON_CLASS(object_class_by_name(apic_type));
+}
+
+void x86_cpu_apic_create(X86CPU *cpu, Error **errp)
+{
+    APICCommonState *apic;
+    ObjectClass *apic_class = OBJECT_CLASS(apic_get_class());
+
+    cpu->apic_state = DEVICE(object_new_with_class(apic_class));
+
+    object_property_add_child(OBJECT(cpu), "lapic",
+                              OBJECT(cpu->apic_state));
+    object_unref(OBJECT(cpu->apic_state));
+
+    qdev_prop_set_uint32(cpu->apic_state, "id", cpu->apic_id);
+    /* TODO: convert to link<> */
+    apic = APIC_COMMON(cpu->apic_state);
+    apic->cpu = cpu;
+    apic->apicbase = APIC_DEFAULT_ADDRESS | MSR_IA32_APICBASE_ENABLE;
+}
+
+void x86_cpu_apic_realize(X86CPU *cpu, Error **errp)
+{
+    APICCommonState *apic;
+    static bool apic_mmio_map_once;
+
+    if (cpu->apic_state == NULL) {
+        return;
+    }
+    qdev_realize(DEVICE(cpu->apic_state), NULL, errp);
+
+    /* Map APIC MMIO area */
+    apic = APIC_COMMON(cpu->apic_state);
+    if (!apic_mmio_map_once) {
+        memory_region_add_subregion_overlap(get_system_memory(),
+                                            apic->apicbase &
+                                            MSR_IA32_APICBASE_BASE,
+                                            &apic->io_memory,
+                                            0x1000);
+        apic_mmio_map_once = true;
+     }
+}
+
+GuestPanicInformation *x86_cpu_get_crash_info(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+    GuestPanicInformation *panic_info = NULL;
+
+    if (env->features[FEAT_HYPERV_EDX] & HV_GUEST_CRASH_MSR_AVAILABLE) {
+        panic_info = g_malloc0(sizeof(GuestPanicInformation));
+
+        panic_info->type = GUEST_PANIC_INFORMATION_TYPE_HYPER_V;
+
+        assert(HV_CRASH_PARAMS >= 5);
+        panic_info->u.hyper_v.arg1 = env->msr_hv_crash_params[0];
+        panic_info->u.hyper_v.arg2 = env->msr_hv_crash_params[1];
+        panic_info->u.hyper_v.arg3 = env->msr_hv_crash_params[2];
+        panic_info->u.hyper_v.arg4 = env->msr_hv_crash_params[3];
+        panic_info->u.hyper_v.arg5 = env->msr_hv_crash_params[4];
+    }
+
+    return panic_info;
+}
+void x86_cpu_get_crash_info_qom(Object *obj, Visitor *v,
+                                const char *name, void *opaque,
+                                Error **errp)
+{
+    CPUState *cs = CPU(obj);
+    GuestPanicInformation *panic_info;
+
+    if (!cs->crash_occurred) {
+        error_setg(errp, "No crash occured");
+        return;
+    }
+
+    panic_info = x86_cpu_get_crash_info(cs);
+    if (panic_info == NULL) {
+        error_setg(errp, "No crash information");
+        return;
+    }
+
+    visit_type_GuestPanicInformation(v, "crash-information", &panic_info,
+                                     errp);
+    qapi_free_GuestPanicInformation(panic_info);
+}
+
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 8d2e1c3136..f9db351f1f 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -20,35 +20,26 @@
 #include "qemu/osdep.h"
 #include "qemu/units.h"
 #include "qemu/cutils.h"
-#include "qemu/bitops.h"
 #include "qemu/qemu-print.h"
 #include "cpu.h"
 #include "tcg/helper-tcg.h"
-#include "exec/exec-all.h"
-#include "sysemu/kvm.h"
 #include "sysemu/reset.h"
 #include "sysemu/hvf.h"
-#include "sysemu/xen.h"
-#include "sysemu/whpx.h"
 #include "kvm/kvm_i386.h"
 #include "sev_i386.h"
-#include "qemu/module.h"
 #include "qapi/qapi-visit-machine.h"
-#include "qapi/qapi-visit-run-state.h"
-#include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qerror.h"
-#include "qom/qom-qobject.h"
 #include "qapi/qapi-commands-machine-target.h"
 #include "standard-headers/asm-x86/kvm_para.h"
 #include "hw/qdev-properties.h"
 #include "hw/i386/topology.h"
 #ifndef CONFIG_USER_ONLY
 #include "exec/address-spaces.h"
-#include "hw/i386/apic_internal.h"
 #include "hw/boards.h"
 #endif
 
 #include "disas/capstone.h"
+#include "cpu-internal.h"
 
 /* Helpers for building CPUID[2] descriptors: */
 
@@ -663,40 +654,7 @@ void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1,
           CPUID_XSAVE_XSAVEC, CPUID_XSAVE_XSAVES */
 #define TCG_14_0_ECX_FEATURES 0
 
-typedef enum FeatureWordType {
-   CPUID_FEATURE_WORD,
-   MSR_FEATURE_WORD,
-} FeatureWordType;
-
-typedef struct FeatureWordInfo {
-    FeatureWordType type;
-    /* feature flags names are taken from "Intel Processor Identification and
-     * the CPUID Instruction" and AMD's "CPUID Specification".
-     * In cases of disagreement between feature naming conventions,
-     * aliases may be added.
-     */
-    const char *feat_names[64];
-    union {
-        /* If type==CPUID_FEATURE_WORD */
-        struct {
-            uint32_t eax;   /* Input EAX for CPUID */
-            bool needs_ecx; /* CPUID instruction uses ECX as input */
-            uint32_t ecx;   /* Input ECX value for CPUID */
-            int reg;        /* output register (R_* constant) */
-        } cpuid;
-        /* If type==MSR_FEATURE_WORD */
-        struct {
-            uint32_t index;
-        } msr;
-    };
-    uint64_t tcg_features; /* Feature flags supported by TCG */
-    uint64_t unmigratable_flags; /* Feature flags known to be unmigratable */
-    uint64_t migratable_flags; /* Feature flags known to be migratable */
-    /* Features that shouldn't be auto-enabled by "-cpu host" */
-    uint64_t no_autoenable_flags;
-} FeatureWordInfo;
-
-static FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
+FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
     [FEAT_1_EDX] = {
         .type = CPUID_FEATURE_WORD,
         .feat_names = {
@@ -4633,7 +4591,6 @@ static void x86_cpu_parse_featurestr(const char *typename, char *features,
     }
 }
 
-static void x86_cpu_expand_features(X86CPU *cpu, Error **errp);
 static void x86_cpu_filter_features(X86CPU *cpu, bool verbose);
 
 /* Build a list with the name of all features on a feature word array */
@@ -5003,207 +4960,6 @@ static void x86_cpu_load_model(X86CPU *cpu, X86CPUModel *model)
     memset(&env->user_features, 0, sizeof(env->user_features));
 }
 
-#ifndef CONFIG_USER_ONLY
-/* Return a QDict containing keys for all properties that can be included
- * in static expansion of CPU models. All properties set by x86_cpu_load_model()
- * must be included in the dictionary.
- */
-static QDict *x86_cpu_static_props(void)
-{
-    FeatureWord w;
-    int i;
-    static const char *props[] = {
-        "min-level",
-        "min-xlevel",
-        "family",
-        "model",
-        "stepping",
-        "model-id",
-        "vendor",
-        "lmce",
-        NULL,
-    };
-    static QDict *d;
-
-    if (d) {
-        return d;
-    }
-
-    d = qdict_new();
-    for (i = 0; props[i]; i++) {
-        qdict_put_null(d, props[i]);
-    }
-
-    for (w = 0; w < FEATURE_WORDS; w++) {
-        FeatureWordInfo *fi = &feature_word_info[w];
-        int bit;
-        for (bit = 0; bit < 64; bit++) {
-            if (!fi->feat_names[bit]) {
-                continue;
-            }
-            qdict_put_null(d, fi->feat_names[bit]);
-        }
-    }
-
-    return d;
-}
-
-/* Add an entry to @props dict, with the value for property. */
-static void x86_cpu_expand_prop(X86CPU *cpu, QDict *props, const char *prop)
-{
-    QObject *value = object_property_get_qobject(OBJECT(cpu), prop,
-                                                 &error_abort);
-
-    qdict_put_obj(props, prop, value);
-}
-
-/* Convert CPU model data from X86CPU object to a property dictionary
- * that can recreate exactly the same CPU model.
- */
-static void x86_cpu_to_dict(X86CPU *cpu, QDict *props)
-{
-    QDict *sprops = x86_cpu_static_props();
-    const QDictEntry *e;
-
-    for (e = qdict_first(sprops); e; e = qdict_next(sprops, e)) {
-        const char *prop = qdict_entry_key(e);
-        x86_cpu_expand_prop(cpu, props, prop);
-    }
-}
-
-/* Convert CPU model data from X86CPU object to a property dictionary
- * that can recreate exactly the same CPU model, including every
- * writeable QOM property.
- */
-static void x86_cpu_to_dict_full(X86CPU *cpu, QDict *props)
-{
-    ObjectPropertyIterator iter;
-    ObjectProperty *prop;
-
-    object_property_iter_init(&iter, OBJECT(cpu));
-    while ((prop = object_property_iter_next(&iter))) {
-        /* skip read-only or write-only properties */
-        if (!prop->get || !prop->set) {
-            continue;
-        }
-
-        /* "hotplugged" is the only property that is configurable
-         * on the command-line but will be set differently on CPUs
-         * created using "-cpu ... -smp ..." and by CPUs created
-         * on the fly by x86_cpu_from_model() for querying. Skip it.
-         */
-        if (!strcmp(prop->name, "hotplugged")) {
-            continue;
-        }
-        x86_cpu_expand_prop(cpu, props, prop->name);
-    }
-}
-
-static void object_apply_props(Object *obj, QDict *props, Error **errp)
-{
-    const QDictEntry *prop;
-
-    for (prop = qdict_first(props); prop; prop = qdict_next(props, prop)) {
-        if (!object_property_set_qobject(obj, qdict_entry_key(prop),
-                                         qdict_entry_value(prop), errp)) {
-            break;
-        }
-    }
-}
-
-/* Create X86CPU object according to model+props specification */
-static X86CPU *x86_cpu_from_model(const char *model, QDict *props, Error **errp)
-{
-    X86CPU *xc = NULL;
-    X86CPUClass *xcc;
-    Error *err = NULL;
-
-    xcc = X86_CPU_CLASS(cpu_class_by_name(TYPE_X86_CPU, model));
-    if (xcc == NULL) {
-        error_setg(&err, "CPU model '%s' not found", model);
-        goto out;
-    }
-
-    xc = X86_CPU(object_new_with_class(OBJECT_CLASS(xcc)));
-    if (props) {
-        object_apply_props(OBJECT(xc), props, &err);
-        if (err) {
-            goto out;
-        }
-    }
-
-    x86_cpu_expand_features(xc, &err);
-    if (err) {
-        goto out;
-    }
-
-out:
-    if (err) {
-        error_propagate(errp, err);
-        object_unref(OBJECT(xc));
-        xc = NULL;
-    }
-    return xc;
-}
-
-CpuModelExpansionInfo *
-qmp_query_cpu_model_expansion(CpuModelExpansionType type,
-                                                      CpuModelInfo *model,
-                                                      Error **errp)
-{
-    X86CPU *xc = NULL;
-    Error *err = NULL;
-    CpuModelExpansionInfo *ret = g_new0(CpuModelExpansionInfo, 1);
-    QDict *props = NULL;
-    const char *base_name;
-
-    xc = x86_cpu_from_model(model->name,
-                            model->has_props ?
-                                qobject_to(QDict, model->props) :
-                                NULL, &err);
-    if (err) {
-        goto out;
-    }
-
-    props = qdict_new();
-    ret->model = g_new0(CpuModelInfo, 1);
-    ret->model->props = QOBJECT(props);
-    ret->model->has_props = true;
-
-    switch (type) {
-    case CPU_MODEL_EXPANSION_TYPE_STATIC:
-        /* Static expansion will be based on "base" only */
-        base_name = "base";
-        x86_cpu_to_dict(xc, props);
-    break;
-    case CPU_MODEL_EXPANSION_TYPE_FULL:
-        /* As we don't return every single property, full expansion needs
-         * to keep the original model name+props, and add extra
-         * properties on top of that.
-         */
-        base_name = model->name;
-        x86_cpu_to_dict_full(xc, props);
-    break;
-    default:
-        error_setg(&err, "Unsupported expansion type");
-        goto out;
-    }
-
-    x86_cpu_to_dict(xc, props);
-
-    ret->model->name = g_strdup(base_name);
-
-out:
-    object_unref(OBJECT(xc));
-    if (err) {
-        error_propagate(errp, err);
-        qapi_free_CpuModelExpansionInfo(ret);
-        ret = NULL;
-    }
-    return ret;
-}
-#endif  /* !CONFIG_USER_ONLY */
-
 static gchar *x86_gdb_arch_name(CPUState *cs)
 {
 #ifdef TARGET_X86_64
@@ -5278,15 +5034,6 @@ static void x86_register_cpudef_types(X86CPUDefinition *def)
 
 }
 
-#if !defined(CONFIG_USER_ONLY)
-
-void cpu_clear_apic_feature(CPUX86State *env)
-{
-    env->features[FEAT_1_EDX] &= ~CPUID_APIC;
-}
-
-#endif /* !CONFIG_USER_ONLY */
-
 void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                    uint32_t *eax, uint32_t *ebx,
                    uint32_t *ecx, uint32_t *edx)
@@ -5934,20 +5681,6 @@ static void x86_cpu_reset(DeviceState *dev)
 #endif
 }
 
-#ifndef CONFIG_USER_ONLY
-bool cpu_is_bsp(X86CPU *cpu)
-{
-    return cpu_get_apic_base(cpu->apic_state) & MSR_IA32_APICBASE_BSP;
-}
-
-/* TODO: remove me, when reset over QOM tree is implemented */
-static void x86_cpu_machine_reset_cb(void *opaque)
-{
-    X86CPU *cpu = opaque;
-    cpu_reset(CPU(cpu));
-}
-#endif
-
 static void mce_init(X86CPU *cpu)
 {
     CPUX86State *cenv = &cpu->env;
@@ -5965,68 +5698,6 @@ static void mce_init(X86CPU *cpu)
     }
 }
 
-#ifndef CONFIG_USER_ONLY
-APICCommonClass *apic_get_class(void)
-{
-    const char *apic_type = "apic";
-
-    /* TODO: in-kernel irqchip for hvf */
-    if (kvm_apic_in_kernel()) {
-        apic_type = "kvm-apic";
-    } else if (xen_enabled()) {
-        apic_type = "xen-apic";
-    } else if (whpx_apic_in_platform()) {
-        apic_type = "whpx-apic";
-    }
-
-    return APIC_COMMON_CLASS(object_class_by_name(apic_type));
-}
-
-static void x86_cpu_apic_create(X86CPU *cpu, Error **errp)
-{
-    APICCommonState *apic;
-    ObjectClass *apic_class = OBJECT_CLASS(apic_get_class());
-
-    cpu->apic_state = DEVICE(object_new_with_class(apic_class));
-
-    object_property_add_child(OBJECT(cpu), "lapic",
-                              OBJECT(cpu->apic_state));
-    object_unref(OBJECT(cpu->apic_state));
-
-    qdev_prop_set_uint32(cpu->apic_state, "id", cpu->apic_id);
-    /* TODO: convert to link<> */
-    apic = APIC_COMMON(cpu->apic_state);
-    apic->cpu = cpu;
-    apic->apicbase = APIC_DEFAULT_ADDRESS | MSR_IA32_APICBASE_ENABLE;
-}
-
-static void x86_cpu_apic_realize(X86CPU *cpu, Error **errp)
-{
-    APICCommonState *apic;
-    static bool apic_mmio_map_once;
-
-    if (cpu->apic_state == NULL) {
-        return;
-    }
-    qdev_realize(DEVICE(cpu->apic_state), NULL, errp);
-
-    /* Map APIC MMIO area */
-    apic = APIC_COMMON(cpu->apic_state);
-    if (!apic_mmio_map_once) {
-        memory_region_add_subregion_overlap(get_system_memory(),
-                                            apic->apicbase &
-                                            MSR_IA32_APICBASE_BASE,
-                                            &apic->io_memory,
-                                            0x1000);
-        apic_mmio_map_once = true;
-     }
-}
-#else
-static void x86_cpu_apic_realize(X86CPU *cpu, Error **errp)
-{
-}
-#endif
-
 static void x86_cpu_adjust_level(X86CPU *cpu, uint32_t *min, uint32_t value)
 {
     if (*min < value) {
@@ -6130,7 +5801,7 @@ static void x86_cpu_enable_xsave_components(X86CPU *cpu)
 /* Expand CPU configuration data, based on configured features
  * and host/accelerator capabilities when appropriate.
  */
-static void x86_cpu_expand_features(X86CPU *cpu, Error **errp)
+void x86_cpu_expand_features(X86CPU *cpu, Error **errp)
 {
     CPUX86State *env = &cpu->env;
     FeatureWord w;
@@ -6502,10 +6173,12 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
             ht_warned = true;
     }
 
+#ifndef CONFIG_USER_ONLY
     x86_cpu_apic_realize(cpu, &local_err);
     if (local_err != NULL) {
         goto out;
     }
+#endif /* !CONFIG_USER_ONLY */
     cpu_reset(cs);
 
     xcc->parent_realize(dev, &local_err);
@@ -6629,52 +6302,6 @@ static void x86_cpu_register_feature_bit_props(X86CPUClass *xcc,
     x86_cpu_register_bit_prop(xcc, name, w, bitnr);
 }
 
-#if !defined(CONFIG_USER_ONLY)
-static GuestPanicInformation *x86_cpu_get_crash_info(CPUState *cs)
-{
-    X86CPU *cpu = X86_CPU(cs);
-    CPUX86State *env = &cpu->env;
-    GuestPanicInformation *panic_info = NULL;
-
-    if (env->features[FEAT_HYPERV_EDX] & HV_GUEST_CRASH_MSR_AVAILABLE) {
-        panic_info = g_malloc0(sizeof(GuestPanicInformation));
-
-        panic_info->type = GUEST_PANIC_INFORMATION_TYPE_HYPER_V;
-
-        assert(HV_CRASH_PARAMS >= 5);
-        panic_info->u.hyper_v.arg1 = env->msr_hv_crash_params[0];
-        panic_info->u.hyper_v.arg2 = env->msr_hv_crash_params[1];
-        panic_info->u.hyper_v.arg3 = env->msr_hv_crash_params[2];
-        panic_info->u.hyper_v.arg4 = env->msr_hv_crash_params[3];
-        panic_info->u.hyper_v.arg5 = env->msr_hv_crash_params[4];
-    }
-
-    return panic_info;
-}
-static void x86_cpu_get_crash_info_qom(Object *obj, Visitor *v,
-                                       const char *name, void *opaque,
-                                       Error **errp)
-{
-    CPUState *cs = CPU(obj);
-    GuestPanicInformation *panic_info;
-
-    if (!cs->crash_occurred) {
-        error_setg(errp, "No crash occured");
-        return;
-    }
-
-    panic_info = x86_cpu_get_crash_info(cs);
-    if (panic_info == NULL) {
-        error_setg(errp, "No crash information");
-        return;
-    }
-
-    visit_type_GuestPanicInformation(v, "crash-information", &panic_info,
-                                     errp);
-    qapi_free_GuestPanicInformation(panic_info);
-}
-#endif /* !CONFIG_USER_ONLY */
-
 static void x86_cpu_initfn(Object *obj)
 {
     X86CPU *cpu = X86_CPU(obj);
diff --git a/target/i386/meson.build b/target/i386/meson.build
index cac26a4581..c2c3cec884 100644
--- a/target/i386/meson.build
+++ b/target/i386/meson.build
@@ -18,6 +18,7 @@ i386_softmmu_ss.add(files(
   'arch_memory_mapping.c',
   'machine.c',
   'monitor.c',
+  'cpu-softmmu.c',
 ))
 i386_user_ss = ss.source_set()
 
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 33+ messages in thread

* Re: [RFC v18 00/15] i386 cleanup PART 2
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (14 preceding siblings ...)
  2021-02-12 12:36 ` [RFC v18 15/15] i386: split off softmmu part of cpu.c Claudio Fontana
@ 2021-02-12 12:57 ` no-reply
  2021-02-15 11:37 ` Alex Bennée
  16 siblings, 0 replies; 33+ messages in thread
From: no-reply @ 2021-02-12 12:57 UTC (permalink / raw)
  To: cfontana
  Cc: lvivier, peter.maydell, thuth, ehabkost, philmd,
	richard.henderson, qemu-devel, r.bolshakov, cfontana, pbonzini,
	alex.bennee

Patchew URL: https://patchew.org/QEMU/20210212123622.15834-1-cfontana@suse.de/



Hi,

This series seems to have some coding style problems. See output below for
more information:

Type: series
Message-id: 20210212123622.15834-1-cfontana@suse.de
Subject: [RFC v18 00/15] i386 cleanup PART 2

=== TEST SCRIPT BEGIN ===
#!/bin/bash
git rev-parse base > /dev/null || exit 0
git config --local diff.renamelimit 0
git config --local diff.renames True
git config --local diff.algorithm histogram
./scripts/checkpatch.pl --mailback base..
=== TEST SCRIPT END ===

Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384
From https://github.com/patchew-project/qemu
 - [tag update]      patchew/20210211232835.2608059-1-crosa@redhat.com -> patchew/20210211232835.2608059-1-crosa@redhat.com
 * [new tag]         patchew/20210212123622.15834-1-cfontana@suse.de -> patchew/20210212123622.15834-1-cfontana@suse.de
Switched to a new branch 'test'
ed48894 i386: split off softmmu part of cpu.c
dd2e304 i386: split seg_helper into user-only and softmmu parts
3e629d7 i386: slit svm_helper into softmmu and stub-only user
cb66e6a i386: separate fpu_helper into user and softmmu parts
1f6097b i386: split misc helper into user and softmmu parts
510ce76 i386: split tcg btp_helper into softmmu and user parts
a3e6766 i386: split tcg excp_helper into softmmu and user parts
d4748e9 i386: split smm helper (softmmu)
9048a77 i386: split off softmmu-only functionality in tcg-cpu
5ebedb2 meson: add target_user_arch
32174a9 accel-cpu: make cpu_realizefn return a bool
4d3e346 target/i386: fix host_cpu_adjust_phys_bits error handling
d14cb3c accel: introduce new accessor functions
f121c98 cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn
c63d0f7 i386: split cpu accelerators from cpu.c, using AccelCPUClass

=== OUTPUT BEGIN ===
1/15 Checking commit c63d0f75db32 (i386: split cpu accelerators from cpu.c, using AccelCPUClass)
WARNING: line over 80 characters
#1335: FILE: target/i386/tcg/tcg-cpu.c:125:
+    memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);

total: 0 errors, 1 warnings, 1256 lines checked

Patch 1/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
2/15 Checking commit f121c98d2a7b (cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn)
3/15 Checking commit d14cb3c31355 (accel: introduce new accessor functions)
4/15 Checking commit 4d3e3462a311 (target/i386: fix host_cpu_adjust_phys_bits error handling)
5/15 Checking commit 32174a937139 (accel-cpu: make cpu_realizefn return a bool)
6/15 Checking commit 5ebedb2a2fe7 (meson: add target_user_arch)
7/15 Checking commit 9048a775db0f (i386: split off softmmu-only functionality in tcg-cpu)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#30: 
new file mode 100644

WARNING: line over 80 characters
#115: FILE: target/i386/tcg/softmmu/tcg-cpu.c:72:
+    memory_region_add_subregion_overlap(cpu->cpu_as_root, 0, cpu->cpu_as_mem, 0);

total: 0 errors, 2 warnings, 212 lines checked

Patch 7/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
8/15 Checking commit d4748e97ef3d (i386: split smm helper (softmmu))
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#18: 
 target/i386/tcg/{ => softmmu}/smm_helper.c | 19 ++-----------------

total: 0 errors, 1 warnings, 77 lines checked

Patch 8/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
9/15 Checking commit a3e67665263e (i386: split tcg excp_helper into softmmu and user parts)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#598: 
new file mode 100644

WARNING: Block comments use a leading /* on a separate line
#736: FILE: target/i386/tcg/softmmu/excp_helper.c:134:
+            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.

WARNING: Block comments use a leading /* on a separate line
#810: FILE: target/i386/tcg/softmmu/excp_helper.c:208:
+/* return value:

WARNING: line over 80 characters
#911: FILE: target/i386/tcg/softmmu/excp_helper.c:309:
+            pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) &

WARNING: Block comments use a leading /* on a separate line
#1005: FILE: target/i386/tcg/softmmu/excp_helper.c:403:
+            /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved.

WARNING: Block comments use a leading /* on a separate line
#1100: FILE: target/i386/tcg/softmmu/excp_helper.c:498:
+        /* only set write access if already dirty... otherwise wait

WARNING: Block comments use * on subsequent lines
#1101: FILE: target/i386/tcg/softmmu/excp_helper.c:499:
+        /* only set write access if already dirty... otherwise wait
+           for dirty access */

WARNING: Block comments use a trailing */ on a separate line
#1101: FILE: target/i386/tcg/softmmu/excp_helper.c:499:
+           for dirty access */

WARNING: Block comments use a leading /* on a separate line
#1114: FILE: target/i386/tcg/softmmu/excp_helper.c:512:
+    /* Even if 4MB pages, we map only one 4KB page in the cache to

WARNING: Block comments use * on subsequent lines
#1115: FILE: target/i386/tcg/softmmu/excp_helper.c:513:
+    /* Even if 4MB pages, we map only one 4KB page in the cache to
+       avoid filling it too fast */

WARNING: Block comments use a trailing */ on a separate line
#1115: FILE: target/i386/tcg/softmmu/excp_helper.c:513:
+       avoid filling it too fast */

ERROR: braces {} are necessary for all arms of this statement
#1129: FILE: target/i386/tcg/softmmu/excp_helper.c:527:
+    if (is_user)
[...]

total: 1 errors, 11 warnings, 612 lines checked

Patch 9/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.

10/15 Checking commit 510ce76bff6f (i386: split tcg btp_helper into softmmu and user parts)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#359: 
new file mode 100644

WARNING: Block comments use a leading /* on a separate line
#495: FILE: target/i386/tcg/softmmu/bpt_helper.c:132:
+    /* If nothing is changing except the global/local enable bits,

WARNING: Block comments use * on subsequent lines
#496: FILE: target/i386/tcg/softmmu/bpt_helper.c:133:
+    /* If nothing is changing except the global/local enable bits,
+       then we can make the change more efficient.  */

WARNING: Block comments use a trailing */ on a separate line
#496: FILE: target/i386/tcg/softmmu/bpt_helper.c:133:
+       then we can make the change more efficient.  */

WARNING: Block comments use a leading /* on a separate line
#498: FILE: target/i386/tcg/softmmu/bpt_helper.c:135:
+        /* Fold the global and local enable bits together into the

WARNING: Block comments use * on subsequent lines
#499: FILE: target/i386/tcg/softmmu/bpt_helper.c:136:
+        /* Fold the global and local enable bits together into the
+           global fields, then xor to show which registers have

WARNING: Block comments use a trailing */ on a separate line
#500: FILE: target/i386/tcg/softmmu/bpt_helper.c:137:
+           changed collective enable state.  */

total: 0 errors, 7 warnings, 616 lines checked

Patch 10/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
11/15 Checking commit 1f6097bf8134 (i386: split misc helper into user and softmmu parts)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#526: 
new file mode 100644

ERROR: switch and case should be at the same indent
#819: FILE: target/i386/tcg/softmmu/misc_helper.c:289:
+    switch ((uint32_t)env->regs[R_ECX]) {
[...]
+     case MSR_IA32_UCODE_REV:

total: 1 errors, 1 warnings, 1009 lines checked

Patch 11/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.

12/15 Checking commit cb66e6ae7b97 (i386: separate fpu_helper into user and softmmu parts)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#142: 
new file mode 100644

total: 0 errors, 1 warnings, 219 lines checked

Patch 12/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
13/15 Checking commit 3e629d77855d (i386: slit svm_helper into softmmu and stub-only user)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#12: 
 target/i386/tcg/{ => softmmu}/svm_helper.c | 62 +-----------------------

total: 0 errors, 1 warnings, 169 lines checked

Patch 13/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
14/15 Checking commit dd2e304724ac (i386: split seg_helper into user-only and softmmu parts)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#320: 
new file mode 100644

ERROR: do not use C99 // comments
#347: FILE: target/i386/tcg/seg_helper.h:23:
+//#define DEBUG_PCALL

WARNING: Block comments use a leading /* on a separate line
#620: FILE: target/i386/tcg/user/seg_helper.c:73:
+    /* Since we emulate only user space, we cannot do more than

WARNING: Block comments use * on subsequent lines
#621: FILE: target/i386/tcg/user/seg_helper.c:74:
+    /* Since we emulate only user space, we cannot do more than
+       exiting the emulation with the suitable exception and error

WARNING: Block comments use a trailing */ on a separate line
#622: FILE: target/i386/tcg/user/seg_helper.c:75:
+       code. So update EIP for INT 0x80 and EXCP_SYSCALL. */

WARNING: Block comments use a leading /* on a separate line
#633: FILE: target/i386/tcg/user/seg_helper.c:86:
+    /* if user mode only, we simulate a fake exception

WARNING: Block comments use * on subsequent lines
#634: FILE: target/i386/tcg/user/seg_helper.c:87:
+    /* if user mode only, we simulate a fake exception
+       which will be handled outside the cpu execution

WARNING: Block comments use a trailing */ on a separate line
#635: FILE: target/i386/tcg/user/seg_helper.c:88:
+       loop */

total: 1 errors, 7 warnings, 595 lines checked

Patch 14/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.

15/15 Checking commit ed4889490139 (i386: split off softmmu part of cpu.c)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating?
#17: 
new file mode 100644

WARNING: Block comments use a leading /* on a separate line
#51: FILE: target/i386/cpu-internal.h:30:
+    /* feature flags names are taken from "Intel Processor Identification and

WARNING: Block comments use a leading /* on a separate line
#134: FILE: target/i386/cpu-softmmu.c:37:
+/* Return a QDict containing keys for all properties that can be included

WARNING: Block comments use a leading /* on a separate line
#187: FILE: target/i386/cpu-softmmu.c:90:
+/* Convert CPU model data from X86CPU object to a property dictionary

WARNING: Block comments use a leading /* on a separate line
#201: FILE: target/i386/cpu-softmmu.c:104:
+/* Convert CPU model data from X86CPU object to a property dictionary

WARNING: Block comments use a leading /* on a separate line
#217: FILE: target/i386/cpu-softmmu.c:120:
+        /* "hotplugged" is the only property that is configurable

WARNING: Block comments use a leading /* on a separate line
#307: FILE: target/i386/cpu-softmmu.c:210:
+        /* As we don't return every single property, full expansion needs

total: 0 errors, 7 warnings, 895 lines checked

Patch 15/15 has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
=== OUTPUT END ===

Test command exited with code: 1


The full log is available at
http://patchew.org/logs/20210212123622.15834-1-cfontana@suse.de/testing.checkpatch/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to patchew-devel@redhat.com

^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 12/15] i386: separate fpu_helper into user and softmmu parts
  2021-02-12 12:36 ` [RFC v18 12/15] i386: separate fpu_helper " Claudio Fontana
@ 2021-02-15 10:32   ` Alex Bennée
  0 siblings, 0 replies; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 10:32 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> Signed-off-by: Claudio Fontana <cfontana@suse.de>

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass
  2021-02-12 12:36 ` [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass Claudio Fontana
@ 2021-02-15 11:29   ` Alex Bennée
  0 siblings, 0 replies; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 11:29 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> i386 is the first user of AccelCPUClass, allowing to split
> cpu.c into:
>
> cpu.c            cpuid and common x86 cpu functionality
> host-cpu.c       host x86 cpu functions and "host" cpu type
> kvm/kvm-cpu.c    KVM x86 AccelCPUClass
> hvf/hvf-cpu.c    HVF x86 AccelCPUClass
> tcg/tcg-cpu.c    TCG x86 AccelCPUClass
>
> Signed-off-by: Claudio Fontana <cfontana@suse.de>

Seems reasonable to me:

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn
  2021-02-12 12:36 ` [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn Claudio Fontana
@ 2021-02-15 11:30   ` Alex Bennée
  0 siblings, 0 replies; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 11:30 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> move the call to accel_cpu->cpu_realizefn to the general
> cpu_exec_realizefn from target/i386, so it does not need to be
> called for every target explicitly as we enable more targets.
>
> Signed-off-by: Claudio Fontana <cfontana@suse.de>

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 03/15] accel: introduce new accessor functions
  2021-02-12 12:36 ` [RFC v18 03/15] accel: introduce new accessor functions Claudio Fontana
@ 2021-02-15 11:34   ` Alex Bennée
  0 siblings, 0 replies; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 11:34 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> avoid open coding the accesses to cpu->accel_cpu interfaces,
> and instead introduce:
>
> accel_cpu_instance_init,
> accel_cpu_realizefn
>
> to be used by the targets/ initfn code,
> and by cpu_exec_realizefn respectively.
>
> Signed-off-by: Claudio Fontana <cfontana@suse.de>

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 06/15] meson: add target_user_arch
  2021-02-12 12:36 ` [RFC v18 06/15] meson: add target_user_arch Claudio Fontana
@ 2021-02-15 11:37   ` Alex Bennée
  0 siblings, 0 replies; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 11:37 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> the lack of target_user_arch makes it hard to fully leverage the
> build system in order to separate user code from softmmu code.
>
> Provide it, so that we can avoid the proliferation of #ifdef
> in target code.
>
> Signed-off-by: Claudio Fontana <cfontana@suse.de>

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 00/15] i386 cleanup PART 2
  2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
                   ` (15 preceding siblings ...)
  2021-02-12 12:57 ` [RFC v18 00/15] i386 cleanup PART 2 no-reply
@ 2021-02-15 11:37 ` Alex Bennée
  2021-02-15 11:48   ` Claudio Fontana
  16 siblings, 1 reply; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 11:37 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

<snip>
> Looking forward to your comments on this proposal,
>
<snip>

So I've reviewed as much as I'm comfortable with - I'm going to defer to
the x86 experts on the split of stuff for x86. However from my point of
view I think it's a nice step in improving modularity and reducing the
maze of #ifdefs in the code.

-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 00/15] i386 cleanup PART 2
  2021-02-15 11:37 ` Alex Bennée
@ 2021-02-15 11:48   ` Claudio Fontana
  0 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 11:48 UTC (permalink / raw)
  To: Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé

On 2/15/21 12:37 PM, Alex Bennée wrote:
> 
> Claudio Fontana <cfontana@suse.de> writes:
> 
> <snip>
>> Looking forward to your comments on this proposal,
>>
> <snip>
> 
> So I've reviewed as much as I'm comfortable with - I'm going to defer to
> the x86 experts on the split of stuff for x86. However from my point of
> view I think it's a nice step in improving modularity and reducing the
> maze of #ifdefs in the code.
> 

Thanks a lot Alex for your review.

If I can leverage your TCG knowledge a bit more, as I forgot a lot about it,

have you noticed in "i386: split tcg btp_helper into softmmu and user parts"
and "i386: split smm helper (softmmu)"

those new

#ifndef CONFIG_USER_ONLY?

Basically we remove some preamble stuff that would end up calling empty stubs for user mode,
but do you think there could be some other consequence I am not seeing?

Maybe there is even more that could be removed in the code immediately preceding those CONFIG_USER_ONLY?

In particular I am referring to patches:

  i386: split tcg btp_helper into softmmu and user parts
  i386: split smm helper (softmmu)

I am commenting those patches now inline, so that it is easier to see what I am talking about.

Again, thanks a lot!

Claudio


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-12 12:36 ` [RFC v18 08/15] i386: split smm helper (softmmu) Claudio Fontana
@ 2021-02-15 11:51   ` Claudio Fontana
  2021-02-15 12:32     ` Alex Bennée
  0 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 11:51 UTC (permalink / raw)
  To: Alex Bennée, Richard Henderson, Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Paolo Bonzini, Thomas Huth, Roman Bolshakov, qemu-devel

On 2/12/21 1:36 PM, Claudio Fontana wrote:
> smm is only really useful for softmmu, split in two modules
> around the CONFIG_USER_ONLY, in order to remove the ifdef
> and use the build system instead.
> 
> Signed-off-by: Claudio Fontana <cfontana@suse.de>
> ---
>  target/i386/helper.h                       |  4 ++++
>  target/i386/tcg/seg_helper.c               |  2 ++
>  target/i386/tcg/{ => softmmu}/smm_helper.c | 19 ++-----------------
>  target/i386/tcg/translate.c                |  2 ++
>  target/i386/tcg/meson.build                |  1 -
>  target/i386/tcg/softmmu/meson.build        |  1 +
>  6 files changed, 11 insertions(+), 18 deletions(-)
>  rename target/i386/tcg/{ => softmmu}/smm_helper.c (98%)
> 
> diff --git a/target/i386/helper.h b/target/i386/helper.h
> index c2ae2f7e61..8ffda4cdc6 100644
> --- a/target/i386/helper.h
> +++ b/target/i386/helper.h
> @@ -70,7 +70,11 @@ DEF_HELPER_1(clac, void, env)
>  DEF_HELPER_1(stac, void, env)
>  DEF_HELPER_3(boundw, void, env, tl, int)
>  DEF_HELPER_3(boundl, void, env, tl, int)
> +
> +#ifndef CONFIG_USER_ONLY
>  DEF_HELPER_1(rsm, void, env)
> +#endif /* !CONFIG_USER_ONLY */
> +
>  DEF_HELPER_2(into, void, env, int)
>  DEF_HELPER_2(cmpxchg8b_unlocked, void, env, tl)
>  DEF_HELPER_2(cmpxchg8b, void, env, tl)
> diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
> index 180d47f0e9..f0cb1bffe7 100644
> --- a/target/i386/tcg/seg_helper.c
> +++ b/target/i386/tcg/seg_helper.c
> @@ -1351,7 +1351,9 @@ bool x86_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
>      case CPU_INTERRUPT_SMI:
>          cpu_svm_check_intercept_param(env, SVM_EXIT_SMI, 0, 0);
>          cs->interrupt_request &= ~CPU_INTERRUPT_SMI;
> +#ifndef CONFIG_USER_ONLY
>          do_smm_enter(cpu);
> +#endif
>          break;
>      case CPU_INTERRUPT_NMI:
>          cpu_svm_check_intercept_param(env, SVM_EXIT_NMI, 0, 0);
> diff --git a/target/i386/tcg/smm_helper.c b/target/i386/tcg/softmmu/smm_helper.c
> similarity index 98%
> rename from target/i386/tcg/smm_helper.c
> rename to target/i386/tcg/softmmu/smm_helper.c
> index 62d027abd3..ee53b26629 100644
> --- a/target/i386/tcg/smm_helper.c
> +++ b/target/i386/tcg/softmmu/smm_helper.c
> @@ -1,5 +1,5 @@
>  /*
> - *  x86 SMM helpers
> + *  x86 SMM helpers (softmmu-only)
>   *
>   *  Copyright (c) 2003 Fabrice Bellard
>   *
> @@ -18,27 +18,14 @@
>   */
>  
>  #include "qemu/osdep.h"
> -#include "qemu/main-loop.h"
>  #include "cpu.h"
>  #include "exec/helper-proto.h"
>  #include "exec/log.h"
> -#include "helper-tcg.h"
> +#include "tcg/helper-tcg.h"
>  
>  
>  /* SMM support */
>  
> -#if defined(CONFIG_USER_ONLY)
> -
> -void do_smm_enter(X86CPU *cpu)
> -{
> -}
> -
> -void helper_rsm(CPUX86State *env)
> -{
> -}
> -
> -#else
> -
>  #ifdef TARGET_X86_64
>  #define SMM_REVISION_ID 0x00020064
>  #else
> @@ -330,5 +317,3 @@ void helper_rsm(CPUX86State *env)
>      qemu_log_mask(CPU_LOG_INT, "SMM: after RSM\n");
>      log_cpu_state_mask(CPU_LOG_INT, CPU(cpu), CPU_DUMP_CCOP);
>  }
> -
> -#endif /* !CONFIG_USER_ONLY */
> diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
> index af1faf9342..5075ac4830 100644
> --- a/target/i386/tcg/translate.c
> +++ b/target/i386/tcg/translate.c
> @@ -8321,7 +8321,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
>              goto illegal_op;
>          gen_update_cc_op(s);
>          gen_jmp_im(s, s->pc - s->cs_base);
> +#ifndef CONFIG_USER_ONLY
>          gen_helper_rsm(cpu_env);
> +#endif /* CONFIG_USER_ONLY */
>          gen_eob(s);
>          break;

Hello Alex,

this is something I wanted to bring in the foreground:

while before we were generating an empty helper call for CONFIG_USER_ONLY,
now we are not generating anything.



>      case 0x1b8: /* SSE4.2 popcnt */
> diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
> index 68fa0c3187..ec5daa1edc 100644
> --- a/target/i386/tcg/meson.build
> +++ b/target/i386/tcg/meson.build
> @@ -8,7 +8,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
>    'misc_helper.c',
>    'mpx_helper.c',
>    'seg_helper.c',
> -  'smm_helper.c',
>    'svm_helper.c',
>    'tcg-cpu.c',
>    'translate.c'), if_false: files('tcg-stub.c'))
> diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
> index 4ab30cc32e..35ba16dc3d 100644
> --- a/target/i386/tcg/softmmu/meson.build
> +++ b/target/i386/tcg/softmmu/meson.build
> @@ -1,3 +1,4 @@
>  i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
>    'tcg-cpu.c',
> +  'smm_helper.c',
>  ))
> 



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 10/15] i386: split tcg btp_helper into softmmu and user parts
  2021-02-12 12:36 ` [RFC v18 10/15] i386: split tcg btp_helper " Claudio Fontana
@ 2021-02-15 11:55   ` Claudio Fontana
  0 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 11:55 UTC (permalink / raw)
  To: Alex Bennée, Paolo Bonzini, Richard Henderson,
	Philippe Mathieu-Daudé,
	Eduardo Habkost, Peter Maydell
  Cc: Laurent Vivier, Thomas Huth, Roman Bolshakov, qemu-devel

On 2/12/21 1:36 PM, Claudio Fontana wrote:
> Signed-off-by: Claudio Fontana <cfontana@suse.de>
> ---
>  target/i386/helper.h                 |   7 +
>  target/i386/tcg/helper-tcg.h         |   3 +
>  target/i386/tcg/bpt_helper.c         | 276 -------------------------
>  target/i386/tcg/softmmu/bpt_helper.c | 293 +++++++++++++++++++++++++++
>  target/i386/tcg/translate.c          |   4 +
>  target/i386/tcg/softmmu/meson.build  |   1 +
>  6 files changed, 308 insertions(+), 276 deletions(-)
>  create mode 100644 target/i386/tcg/softmmu/bpt_helper.c
> 
> diff --git a/target/i386/helper.h b/target/i386/helper.h
> index 8ffda4cdc6..095520f81f 100644
> --- a/target/i386/helper.h
> +++ b/target/i386/helper.h
> @@ -46,7 +46,11 @@ DEF_HELPER_2(read_crN, tl, env, int)
>  DEF_HELPER_3(write_crN, void, env, int, tl)
>  DEF_HELPER_2(lmsw, void, env, tl)
>  DEF_HELPER_1(clts, void, env)
> +
> +#ifndef CONFIG_USER_ONLY
>  DEF_HELPER_FLAGS_3(set_dr, TCG_CALL_NO_WG, void, env, int, tl)
> +#endif /* !CONFIG_USER_ONLY */
> +
>  DEF_HELPER_FLAGS_2(get_dr, TCG_CALL_NO_WG, tl, env, int)
>  DEF_HELPER_2(invlpg, void, env, tl)
>  
> @@ -100,7 +104,10 @@ DEF_HELPER_3(outw, void, env, i32, i32)
>  DEF_HELPER_2(inw, tl, env, i32)
>  DEF_HELPER_3(outl, void, env, i32, i32)
>  DEF_HELPER_2(inl, tl, env, i32)
> +
> +#ifndef CONFIG_USER_ONLY
>  DEF_HELPER_FLAGS_4(bpt_io, TCG_CALL_NO_WG, void, env, i32, i32, tl)
> +#endif /* !CONFIG_USER_ONLY */
>  
>  DEF_HELPER_3(svm_check_intercept_param, void, env, i32, i64)
>  DEF_HELPER_4(svm_check_io, void, env, i32, i32, i32)
> diff --git a/target/i386/tcg/helper-tcg.h b/target/i386/tcg/helper-tcg.h
> index c133c63555..b420b3356d 100644
> --- a/target/i386/tcg/helper-tcg.h
> +++ b/target/i386/tcg/helper-tcg.h
> @@ -92,4 +92,7 @@ void do_interrupt_x86_hardirq(CPUX86State *env, int intno, int is_hw);
>  /* smm_helper.c */
>  void do_smm_enter(X86CPU *cpu);
>  
> +/* bpt_helper.c */
> +bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update);
> +
>  #endif /* I386_HELPER_TCG_H */
> diff --git a/target/i386/tcg/bpt_helper.c b/target/i386/tcg/bpt_helper.c
> index 979230ac12..fb2a65ac9c 100644
> --- a/target/i386/tcg/bpt_helper.c
> +++ b/target/i386/tcg/bpt_helper.c
> @@ -19,223 +19,9 @@
>  
>  #include "qemu/osdep.h"
>  #include "cpu.h"
> -#include "exec/exec-all.h"
>  #include "exec/helper-proto.h"
>  #include "helper-tcg.h"
>  
> -
> -#ifndef CONFIG_USER_ONLY
> -static inline bool hw_local_breakpoint_enabled(unsigned long dr7, int index)
> -{
> -    return (dr7 >> (index * 2)) & 1;
> -}
> -
> -static inline bool hw_global_breakpoint_enabled(unsigned long dr7, int index)
> -{
> -    return (dr7 >> (index * 2)) & 2;
> -
> -}
> -static inline bool hw_breakpoint_enabled(unsigned long dr7, int index)
> -{
> -    return hw_global_breakpoint_enabled(dr7, index) ||
> -           hw_local_breakpoint_enabled(dr7, index);
> -}
> -
> -static inline int hw_breakpoint_type(unsigned long dr7, int index)
> -{
> -    return (dr7 >> (DR7_TYPE_SHIFT + (index * 4))) & 3;
> -}
> -
> -static inline int hw_breakpoint_len(unsigned long dr7, int index)
> -{
> -    int len = ((dr7 >> (DR7_LEN_SHIFT + (index * 4))) & 3);
> -    return (len == 2) ? 8 : len + 1;
> -}
> -
> -static int hw_breakpoint_insert(CPUX86State *env, int index)
> -{
> -    CPUState *cs = env_cpu(env);
> -    target_ulong dr7 = env->dr[7];
> -    target_ulong drN = env->dr[index];
> -    int err = 0;
> -
> -    switch (hw_breakpoint_type(dr7, index)) {
> -    case DR7_TYPE_BP_INST:
> -        if (hw_breakpoint_enabled(dr7, index)) {
> -            err = cpu_breakpoint_insert(cs, drN, BP_CPU,
> -                                        &env->cpu_breakpoint[index]);
> -        }
> -        break;
> -
> -    case DR7_TYPE_IO_RW:
> -        /* Notice when we should enable calls to bpt_io.  */
> -        return hw_breakpoint_enabled(env->dr[7], index)
> -               ? HF_IOBPT_MASK : 0;
> -
> -    case DR7_TYPE_DATA_WR:
> -        if (hw_breakpoint_enabled(dr7, index)) {
> -            err = cpu_watchpoint_insert(cs, drN,
> -                                        hw_breakpoint_len(dr7, index),
> -                                        BP_CPU | BP_MEM_WRITE,
> -                                        &env->cpu_watchpoint[index]);
> -        }
> -        break;
> -
> -    case DR7_TYPE_DATA_RW:
> -        if (hw_breakpoint_enabled(dr7, index)) {
> -            err = cpu_watchpoint_insert(cs, drN,
> -                                        hw_breakpoint_len(dr7, index),
> -                                        BP_CPU | BP_MEM_ACCESS,
> -                                        &env->cpu_watchpoint[index]);
> -        }
> -        break;
> -    }
> -    if (err) {
> -        env->cpu_breakpoint[index] = NULL;
> -    }
> -    return 0;
> -}
> -
> -static void hw_breakpoint_remove(CPUX86State *env, int index)
> -{
> -    CPUState *cs = env_cpu(env);
> -
> -    switch (hw_breakpoint_type(env->dr[7], index)) {
> -    case DR7_TYPE_BP_INST:
> -        if (env->cpu_breakpoint[index]) {
> -            cpu_breakpoint_remove_by_ref(cs, env->cpu_breakpoint[index]);
> -            env->cpu_breakpoint[index] = NULL;
> -        }
> -        break;
> -
> -    case DR7_TYPE_DATA_WR:
> -    case DR7_TYPE_DATA_RW:
> -        if (env->cpu_breakpoint[index]) {
> -            cpu_watchpoint_remove_by_ref(cs, env->cpu_watchpoint[index]);
> -            env->cpu_breakpoint[index] = NULL;
> -        }
> -        break;
> -
> -    case DR7_TYPE_IO_RW:
> -        /* HF_IOBPT_MASK cleared elsewhere.  */
> -        break;
> -    }
> -}
> -
> -void cpu_x86_update_dr7(CPUX86State *env, uint32_t new_dr7)
> -{
> -    target_ulong old_dr7 = env->dr[7];
> -    int iobpt = 0;
> -    int i;
> -
> -    new_dr7 |= DR7_FIXED_1;
> -
> -    /* If nothing is changing except the global/local enable bits,
> -       then we can make the change more efficient.  */
> -    if (((old_dr7 ^ new_dr7) & ~0xff) == 0) {
> -        /* Fold the global and local enable bits together into the
> -           global fields, then xor to show which registers have
> -           changed collective enable state.  */
> -        int mod = ((old_dr7 | old_dr7 * 2) ^ (new_dr7 | new_dr7 * 2)) & 0xff;
> -
> -        for (i = 0; i < DR7_MAX_BP; i++) {
> -            if ((mod & (2 << i * 2)) && !hw_breakpoint_enabled(new_dr7, i)) {
> -                hw_breakpoint_remove(env, i);
> -            }
> -        }
> -        env->dr[7] = new_dr7;
> -        for (i = 0; i < DR7_MAX_BP; i++) {
> -            if (mod & (2 << i * 2) && hw_breakpoint_enabled(new_dr7, i)) {
> -                iobpt |= hw_breakpoint_insert(env, i);
> -            } else if (hw_breakpoint_type(new_dr7, i) == DR7_TYPE_IO_RW
> -                       && hw_breakpoint_enabled(new_dr7, i)) {
> -                iobpt |= HF_IOBPT_MASK;
> -            }
> -        }
> -    } else {
> -        for (i = 0; i < DR7_MAX_BP; i++) {
> -            hw_breakpoint_remove(env, i);
> -        }
> -        env->dr[7] = new_dr7;
> -        for (i = 0; i < DR7_MAX_BP; i++) {
> -            iobpt |= hw_breakpoint_insert(env, i);
> -        }
> -    }
> -
> -    env->hflags = (env->hflags & ~HF_IOBPT_MASK) | iobpt;
> -}
> -
> -static bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update)
> -{
> -    target_ulong dr6;
> -    int reg;
> -    bool hit_enabled = false;
> -
> -    dr6 = env->dr[6] & ~0xf;
> -    for (reg = 0; reg < DR7_MAX_BP; reg++) {
> -        bool bp_match = false;
> -        bool wp_match = false;
> -
> -        switch (hw_breakpoint_type(env->dr[7], reg)) {
> -        case DR7_TYPE_BP_INST:
> -            if (env->dr[reg] == env->eip) {
> -                bp_match = true;
> -            }
> -            break;
> -        case DR7_TYPE_DATA_WR:
> -        case DR7_TYPE_DATA_RW:
> -            if (env->cpu_watchpoint[reg] &&
> -                env->cpu_watchpoint[reg]->flags & BP_WATCHPOINT_HIT) {
> -                wp_match = true;
> -            }
> -            break;
> -        case DR7_TYPE_IO_RW:
> -            break;
> -        }
> -        if (bp_match || wp_match) {
> -            dr6 |= 1 << reg;
> -            if (hw_breakpoint_enabled(env->dr[7], reg)) {
> -                hit_enabled = true;
> -            }
> -        }
> -    }
> -
> -    if (hit_enabled || force_dr6_update) {
> -        env->dr[6] = dr6;
> -    }
> -
> -    return hit_enabled;
> -}
> -
> -void breakpoint_handler(CPUState *cs)
> -{
> -    X86CPU *cpu = X86_CPU(cs);
> -    CPUX86State *env = &cpu->env;
> -    CPUBreakpoint *bp;
> -
> -    if (cs->watchpoint_hit) {
> -        if (cs->watchpoint_hit->flags & BP_CPU) {
> -            cs->watchpoint_hit = NULL;
> -            if (check_hw_breakpoints(env, false)) {
> -                raise_exception(env, EXCP01_DB);
> -            } else {
> -                cpu_loop_exit_noexc(cs);
> -            }
> -        }
> -    } else {
> -        QTAILQ_FOREACH(bp, &cs->breakpoints, entry) {
> -            if (bp->pc == env->eip) {
> -                if (bp->flags & BP_CPU) {
> -                    check_hw_breakpoints(env, true);
> -                    raise_exception(env, EXCP01_DB);
> -                }
> -                break;
> -            }
> -        }
> -    }
> -}
> -#endif
> -
>  void helper_single_step(CPUX86State *env)
>  {
>  #ifndef CONFIG_USER_ONLY
> @@ -252,41 +38,6 @@ void helper_rechecking_single_step(CPUX86State *env)
>      }
>  }
>  
> -void helper_set_dr(CPUX86State *env, int reg, target_ulong t0)
> -{
> -#ifndef CONFIG_USER_ONLY
> -    switch (reg) {
> -    case 0: case 1: case 2: case 3:
> -        if (hw_breakpoint_enabled(env->dr[7], reg)
> -            && hw_breakpoint_type(env->dr[7], reg) != DR7_TYPE_IO_RW) {
> -            hw_breakpoint_remove(env, reg);
> -            env->dr[reg] = t0;
> -            hw_breakpoint_insert(env, reg);
> -        } else {
> -            env->dr[reg] = t0;
> -        }
> -        return;
> -    case 4:
> -        if (env->cr[4] & CR4_DE_MASK) {
> -            break;
> -        }
> -        /* fallthru */
> -    case 6:
> -        env->dr[6] = t0 | DR6_FIXED_1;
> -        return;
> -    case 5:
> -        if (env->cr[4] & CR4_DE_MASK) {
> -            break;
> -        }
> -        /* fallthru */
> -    case 7:
> -        cpu_x86_update_dr7(env, t0);
> -        return;
> -    }
> -    raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
> -#endif
> -}
> -
>  target_ulong helper_get_dr(CPUX86State *env, int reg)
>  {
>      switch (reg) {
> @@ -307,30 +58,3 @@ target_ulong helper_get_dr(CPUX86State *env, int reg)
>      }
>      raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
>  }
> -
> -/* Check if Port I/O is trapped by a breakpoint.  */
> -void helper_bpt_io(CPUX86State *env, uint32_t port,
> -                   uint32_t size, target_ulong next_eip)
> -{
> -#ifndef CONFIG_USER_ONLY
> -    target_ulong dr7 = env->dr[7];
> -    int i, hit = 0;
> -
> -    for (i = 0; i < DR7_MAX_BP; ++i) {
> -        if (hw_breakpoint_type(dr7, i) == DR7_TYPE_IO_RW
> -            && hw_breakpoint_enabled(dr7, i)) {
> -            int bpt_len = hw_breakpoint_len(dr7, i);
> -            if (port + size - 1 >= env->dr[i]
> -                && port <= env->dr[i] + bpt_len - 1) {
> -                hit |= 1 << i;
> -            }
> -        }
> -    }
> -
> -    if (hit) {
> -        env->dr[6] = (env->dr[6] & ~0xf) | hit;
> -        env->eip = next_eip;
> -        raise_exception(env, EXCP01_DB);
> -    }
> -#endif
> -}
> diff --git a/target/i386/tcg/softmmu/bpt_helper.c b/target/i386/tcg/softmmu/bpt_helper.c
> new file mode 100644
> index 0000000000..1f3dd09cad
> --- /dev/null
> +++ b/target/i386/tcg/softmmu/bpt_helper.c
> @@ -0,0 +1,293 @@
> +/*
> + *  i386 breakpoint helpers - softmmu-only code
> + *
> + *  Copyright (c) 2003 Fabrice Bellard
> + *
> + * This library is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation; either
> + * version 2.1 of the License, or (at your option) any later version.
> + *
> + * This library is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with this library; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include "qemu/osdep.h"
> +#include "cpu.h"
> +#include "exec/exec-all.h"
> +#include "exec/helper-proto.h"
> +#include "tcg/helper-tcg.h"
> +
> +
> +static inline bool hw_local_breakpoint_enabled(unsigned long dr7, int index)
> +{
> +    return (dr7 >> (index * 2)) & 1;
> +}
> +
> +static inline bool hw_global_breakpoint_enabled(unsigned long dr7, int index)
> +{
> +    return (dr7 >> (index * 2)) & 2;
> +
> +}
> +static inline bool hw_breakpoint_enabled(unsigned long dr7, int index)
> +{
> +    return hw_global_breakpoint_enabled(dr7, index) ||
> +           hw_local_breakpoint_enabled(dr7, index);
> +}
> +
> +static inline int hw_breakpoint_type(unsigned long dr7, int index)
> +{
> +    return (dr7 >> (DR7_TYPE_SHIFT + (index * 4))) & 3;
> +}
> +
> +static inline int hw_breakpoint_len(unsigned long dr7, int index)
> +{
> +    int len = ((dr7 >> (DR7_LEN_SHIFT + (index * 4))) & 3);
> +    return (len == 2) ? 8 : len + 1;
> +}
> +
> +static int hw_breakpoint_insert(CPUX86State *env, int index)
> +{
> +    CPUState *cs = env_cpu(env);
> +    target_ulong dr7 = env->dr[7];
> +    target_ulong drN = env->dr[index];
> +    int err = 0;
> +
> +    switch (hw_breakpoint_type(dr7, index)) {
> +    case DR7_TYPE_BP_INST:
> +        if (hw_breakpoint_enabled(dr7, index)) {
> +            err = cpu_breakpoint_insert(cs, drN, BP_CPU,
> +                                        &env->cpu_breakpoint[index]);
> +        }
> +        break;
> +
> +    case DR7_TYPE_IO_RW:
> +        /* Notice when we should enable calls to bpt_io.  */
> +        return hw_breakpoint_enabled(env->dr[7], index)
> +               ? HF_IOBPT_MASK : 0;
> +
> +    case DR7_TYPE_DATA_WR:
> +        if (hw_breakpoint_enabled(dr7, index)) {
> +            err = cpu_watchpoint_insert(cs, drN,
> +                                        hw_breakpoint_len(dr7, index),
> +                                        BP_CPU | BP_MEM_WRITE,
> +                                        &env->cpu_watchpoint[index]);
> +        }
> +        break;
> +
> +    case DR7_TYPE_DATA_RW:
> +        if (hw_breakpoint_enabled(dr7, index)) {
> +            err = cpu_watchpoint_insert(cs, drN,
> +                                        hw_breakpoint_len(dr7, index),
> +                                        BP_CPU | BP_MEM_ACCESS,
> +                                        &env->cpu_watchpoint[index]);
> +        }
> +        break;
> +    }
> +    if (err) {
> +        env->cpu_breakpoint[index] = NULL;
> +    }
> +    return 0;
> +}
> +
> +static void hw_breakpoint_remove(CPUX86State *env, int index)
> +{
> +    CPUState *cs = env_cpu(env);
> +
> +    switch (hw_breakpoint_type(env->dr[7], index)) {
> +    case DR7_TYPE_BP_INST:
> +        if (env->cpu_breakpoint[index]) {
> +            cpu_breakpoint_remove_by_ref(cs, env->cpu_breakpoint[index]);
> +            env->cpu_breakpoint[index] = NULL;
> +        }
> +        break;
> +
> +    case DR7_TYPE_DATA_WR:
> +    case DR7_TYPE_DATA_RW:
> +        if (env->cpu_breakpoint[index]) {
> +            cpu_watchpoint_remove_by_ref(cs, env->cpu_watchpoint[index]);
> +            env->cpu_breakpoint[index] = NULL;
> +        }
> +        break;
> +
> +    case DR7_TYPE_IO_RW:
> +        /* HF_IOBPT_MASK cleared elsewhere.  */
> +        break;
> +    }
> +}
> +
> +void cpu_x86_update_dr7(CPUX86State *env, uint32_t new_dr7)
> +{
> +    target_ulong old_dr7 = env->dr[7];
> +    int iobpt = 0;
> +    int i;
> +
> +    new_dr7 |= DR7_FIXED_1;
> +
> +    /* If nothing is changing except the global/local enable bits,
> +       then we can make the change more efficient.  */
> +    if (((old_dr7 ^ new_dr7) & ~0xff) == 0) {
> +        /* Fold the global and local enable bits together into the
> +           global fields, then xor to show which registers have
> +           changed collective enable state.  */
> +        int mod = ((old_dr7 | old_dr7 * 2) ^ (new_dr7 | new_dr7 * 2)) & 0xff;
> +
> +        for (i = 0; i < DR7_MAX_BP; i++) {
> +            if ((mod & (2 << i * 2)) && !hw_breakpoint_enabled(new_dr7, i)) {
> +                hw_breakpoint_remove(env, i);
> +            }
> +        }
> +        env->dr[7] = new_dr7;
> +        for (i = 0; i < DR7_MAX_BP; i++) {
> +            if (mod & (2 << i * 2) && hw_breakpoint_enabled(new_dr7, i)) {
> +                iobpt |= hw_breakpoint_insert(env, i);
> +            } else if (hw_breakpoint_type(new_dr7, i) == DR7_TYPE_IO_RW
> +                       && hw_breakpoint_enabled(new_dr7, i)) {
> +                iobpt |= HF_IOBPT_MASK;
> +            }
> +        }
> +    } else {
> +        for (i = 0; i < DR7_MAX_BP; i++) {
> +            hw_breakpoint_remove(env, i);
> +        }
> +        env->dr[7] = new_dr7;
> +        for (i = 0; i < DR7_MAX_BP; i++) {
> +            iobpt |= hw_breakpoint_insert(env, i);
> +        }
> +    }
> +
> +    env->hflags = (env->hflags & ~HF_IOBPT_MASK) | iobpt;
> +}
> +
> +bool check_hw_breakpoints(CPUX86State *env, bool force_dr6_update)
> +{
> +    target_ulong dr6;
> +    int reg;
> +    bool hit_enabled = false;
> +
> +    dr6 = env->dr[6] & ~0xf;
> +    for (reg = 0; reg < DR7_MAX_BP; reg++) {
> +        bool bp_match = false;
> +        bool wp_match = false;
> +
> +        switch (hw_breakpoint_type(env->dr[7], reg)) {
> +        case DR7_TYPE_BP_INST:
> +            if (env->dr[reg] == env->eip) {
> +                bp_match = true;
> +            }
> +            break;
> +        case DR7_TYPE_DATA_WR:
> +        case DR7_TYPE_DATA_RW:
> +            if (env->cpu_watchpoint[reg] &&
> +                env->cpu_watchpoint[reg]->flags & BP_WATCHPOINT_HIT) {
> +                wp_match = true;
> +            }
> +            break;
> +        case DR7_TYPE_IO_RW:
> +            break;
> +        }
> +        if (bp_match || wp_match) {
> +            dr6 |= 1 << reg;
> +            if (hw_breakpoint_enabled(env->dr[7], reg)) {
> +                hit_enabled = true;
> +            }
> +        }
> +    }
> +
> +    if (hit_enabled || force_dr6_update) {
> +        env->dr[6] = dr6;
> +    }
> +
> +    return hit_enabled;
> +}
> +
> +void breakpoint_handler(CPUState *cs)
> +{
> +    X86CPU *cpu = X86_CPU(cs);
> +    CPUX86State *env = &cpu->env;
> +    CPUBreakpoint *bp;
> +
> +    if (cs->watchpoint_hit) {
> +        if (cs->watchpoint_hit->flags & BP_CPU) {
> +            cs->watchpoint_hit = NULL;
> +            if (check_hw_breakpoints(env, false)) {
> +                raise_exception(env, EXCP01_DB);
> +            } else {
> +                cpu_loop_exit_noexc(cs);
> +            }
> +        }
> +    } else {
> +        QTAILQ_FOREACH(bp, &cs->breakpoints, entry) {
> +            if (bp->pc == env->eip) {
> +                if (bp->flags & BP_CPU) {
> +                    check_hw_breakpoints(env, true);
> +                    raise_exception(env, EXCP01_DB);
> +                }
> +                break;
> +            }
> +        }
> +    }
> +}
> +
> +void helper_set_dr(CPUX86State *env, int reg, target_ulong t0)
> +{
> +    switch (reg) {
> +    case 0: case 1: case 2: case 3:
> +        if (hw_breakpoint_enabled(env->dr[7], reg)
> +            && hw_breakpoint_type(env->dr[7], reg) != DR7_TYPE_IO_RW) {
> +            hw_breakpoint_remove(env, reg);
> +            env->dr[reg] = t0;
> +            hw_breakpoint_insert(env, reg);
> +        } else {
> +            env->dr[reg] = t0;
> +        }
> +        return;
> +    case 4:
> +        if (env->cr[4] & CR4_DE_MASK) {
> +            break;
> +        }
> +        /* fallthru */
> +    case 6:
> +        env->dr[6] = t0 | DR6_FIXED_1;
> +        return;
> +    case 5:
> +        if (env->cr[4] & CR4_DE_MASK) {
> +            break;
> +        }
> +        /* fallthru */
> +    case 7:
> +        cpu_x86_update_dr7(env, t0);
> +        return;
> +    }
> +    raise_exception_err_ra(env, EXCP06_ILLOP, 0, GETPC());
> +}
> +
> +/* Check if Port I/O is trapped by a breakpoint.  */
> +void helper_bpt_io(CPUX86State *env, uint32_t port,
> +                   uint32_t size, target_ulong next_eip)
> +{
> +    target_ulong dr7 = env->dr[7];
> +    int i, hit = 0;
> +
> +    for (i = 0; i < DR7_MAX_BP; ++i) {
> +        if (hw_breakpoint_type(dr7, i) == DR7_TYPE_IO_RW
> +            && hw_breakpoint_enabled(dr7, i)) {
> +            int bpt_len = hw_breakpoint_len(dr7, i);
> +            if (port + size - 1 >= env->dr[i]
> +                && port <= env->dr[i] + bpt_len - 1) {
> +                hit |= 1 << i;
> +            }
> +        }
> +    }
> +
> +    if (hit) {
> +        env->dr[6] = (env->dr[6] & ~0xf) | hit;
> +        env->eip = next_eip;
> +        raise_exception(env, EXCP01_DB);
> +    }
> +}
> diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
> index 5075ac4830..35f5d5c73e 100644
> --- a/target/i386/tcg/translate.c
> +++ b/target/i386/tcg/translate.c
> @@ -1116,6 +1116,7 @@ static inline void gen_cmps(DisasContext *s, MemOp ot)
>  
>  static void gen_bpt_io(DisasContext *s, TCGv_i32 t_port, int ot)
>  {
> +#ifndef CONFIG_USER_ONLY
>      if (s->flags & HF_IOBPT_MASK) {
>          TCGv_i32 t_size = tcg_const_i32(1 << ot);
>          TCGv t_next = tcg_const_tl(s->pc - s->cs_base);
> @@ -1124,6 +1125,7 @@ static void gen_bpt_io(DisasContext *s, TCGv_i32 t_port, int ot)
>          tcg_temp_free_i32(t_size);
>          tcg_temp_free(t_next);
>      }
> +#endif /* !CONFIG_USER_ONLY */
>  }
>  
>  
> @@ -8074,7 +8076,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
>                  gen_svm_check_intercept(s, pc_start, SVM_EXIT_WRITE_DR0 + reg);
>                  gen_op_mov_v_reg(s, ot, s->T0, rm);
>                  tcg_gen_movi_i32(s->tmp2_i32, reg);
> +#ifndef CONFIG_USER_ONLY
>                  gen_helper_set_dr(cpu_env, s->tmp2_i32, s->T0);
> +#endif /* CONFIG_USER_ONLY */


Hi Alex,

here is another case I would like to bring to the foreground,

also here for CONFIG_USER_ONLY we do not generate anything, while before we were generating a call to an empty helper.

In this case, do you think that more should be reworked in the instructions before?
Are those only present in order to create the parameters to gen_helper_set_dr, and should therefore be wrapped in the #ifndef as well?

Thanks,

Claudio


>                  gen_jmp_im(s, s->pc - s->cs_base);
>                  gen_eob(s);
>              } else {
> diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
> index 6d0a0a0fee..1580950141 100644
> --- a/target/i386/tcg/softmmu/meson.build
> +++ b/target/i386/tcg/softmmu/meson.build
> @@ -2,4 +2,5 @@ i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
>    'tcg-cpu.c',
>    'smm_helper.c',
>    'excp_helper.c',
> +  'bpt_helper.c',
>  ))
> 



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 11:51   ` Claudio Fontana
@ 2021-02-15 12:32     ` Alex Bennée
  2021-02-15 12:59       ` Claudio Fontana
  0 siblings, 1 reply; 33+ messages in thread
From: Alex Bennée @ 2021-02-15 12:32 UTC (permalink / raw)
  To: Claudio Fontana
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé


Claudio Fontana <cfontana@suse.de> writes:

> On 2/12/21 1:36 PM, Claudio Fontana wrote:
>> smm is only really useful for softmmu, split in two modules
>> around the CONFIG_USER_ONLY, in order to remove the ifdef
>> and use the build system instead.
>> 
>> Signed-off-by: Claudio Fontana <cfontana@suse.de>
>> ---
>>  target/i386/helper.h                       |  4 ++++
>>  target/i386/tcg/seg_helper.c               |  2 ++
>>  target/i386/tcg/{ => softmmu}/smm_helper.c | 19 ++-----------------
>>  target/i386/tcg/translate.c                |  2 ++
>>  target/i386/tcg/meson.build                |  1 -
>>  target/i386/tcg/softmmu/meson.build        |  1 +
>>  6 files changed, 11 insertions(+), 18 deletions(-)
>>  rename target/i386/tcg/{ => softmmu}/smm_helper.c (98%)
>> 
>> diff --git a/target/i386/helper.h b/target/i386/helper.h
>> index c2ae2f7e61..8ffda4cdc6 100644
>> --- a/target/i386/helper.h
>> +++ b/target/i386/helper.h
>> @@ -70,7 +70,11 @@ DEF_HELPER_1(clac, void, env)
>>  DEF_HELPER_1(stac, void, env)
>>  DEF_HELPER_3(boundw, void, env, tl, int)
>>  DEF_HELPER_3(boundl, void, env, tl, int)
>> +
>> +#ifndef CONFIG_USER_ONLY
>>  DEF_HELPER_1(rsm, void, env)
>> +#endif /* !CONFIG_USER_ONLY */
>> +
>>  DEF_HELPER_2(into, void, env, int)
>>  DEF_HELPER_2(cmpxchg8b_unlocked, void, env, tl)
>>  DEF_HELPER_2(cmpxchg8b, void, env, tl)
>> diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
>> index 180d47f0e9..f0cb1bffe7 100644
>> --- a/target/i386/tcg/seg_helper.c
>> +++ b/target/i386/tcg/seg_helper.c
>> @@ -1351,7 +1351,9 @@ bool x86_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
>>      case CPU_INTERRUPT_SMI:
>>          cpu_svm_check_intercept_param(env, SVM_EXIT_SMI, 0, 0);
>>          cs->interrupt_request &= ~CPU_INTERRUPT_SMI;
>> +#ifndef CONFIG_USER_ONLY
>>          do_smm_enter(cpu);
>> +#endif
>>          break;
>>      case CPU_INTERRUPT_NMI:
>>          cpu_svm_check_intercept_param(env, SVM_EXIT_NMI, 0, 0);
>> diff --git a/target/i386/tcg/smm_helper.c b/target/i386/tcg/softmmu/smm_helper.c
>> similarity index 98%
>> rename from target/i386/tcg/smm_helper.c
>> rename to target/i386/tcg/softmmu/smm_helper.c
>> index 62d027abd3..ee53b26629 100644
>> --- a/target/i386/tcg/smm_helper.c
>> +++ b/target/i386/tcg/softmmu/smm_helper.c
>> @@ -1,5 +1,5 @@
>>  /*
>> - *  x86 SMM helpers
>> + *  x86 SMM helpers (softmmu-only)
>>   *
>>   *  Copyright (c) 2003 Fabrice Bellard
>>   *
>> @@ -18,27 +18,14 @@
>>   */
>>  
>>  #include "qemu/osdep.h"
>> -#include "qemu/main-loop.h"
>>  #include "cpu.h"
>>  #include "exec/helper-proto.h"
>>  #include "exec/log.h"
>> -#include "helper-tcg.h"
>> +#include "tcg/helper-tcg.h"
>>  
>>  
>>  /* SMM support */
>>  
>> -#if defined(CONFIG_USER_ONLY)
>> -
>> -void do_smm_enter(X86CPU *cpu)
>> -{
>> -}
>> -
>> -void helper_rsm(CPUX86State *env)
>> -{
>> -}
>> -
>> -#else
>> -
>>  #ifdef TARGET_X86_64
>>  #define SMM_REVISION_ID 0x00020064
>>  #else
>> @@ -330,5 +317,3 @@ void helper_rsm(CPUX86State *env)
>>      qemu_log_mask(CPU_LOG_INT, "SMM: after RSM\n");
>>      log_cpu_state_mask(CPU_LOG_INT, CPU(cpu), CPU_DUMP_CCOP);
>>  }
>> -
>> -#endif /* !CONFIG_USER_ONLY */
>> diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
>> index af1faf9342..5075ac4830 100644
>> --- a/target/i386/tcg/translate.c
>> +++ b/target/i386/tcg/translate.c
>> @@ -8321,7 +8321,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
>>              goto illegal_op;
>>          gen_update_cc_op(s);
>>          gen_jmp_im(s, s->pc - s->cs_base);
>> +#ifndef CONFIG_USER_ONLY
>>          gen_helper_rsm(cpu_env);
>> +#endif /* CONFIG_USER_ONLY */
>>          gen_eob(s);
>>          break;
>
> Hello Alex,
>
> this is something I wanted to bring in the foreground:
>
> while before we were generating an empty helper call for CONFIG_USER_ONLY,
> now we are not generating anything.

Surely that says we only generate the helper call when we are not
CONFIG_USER_ONLY?


>
>
>
>>      case 0x1b8: /* SSE4.2 popcnt */
>> diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
>> index 68fa0c3187..ec5daa1edc 100644
>> --- a/target/i386/tcg/meson.build
>> +++ b/target/i386/tcg/meson.build
>> @@ -8,7 +8,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
>>    'misc_helper.c',
>>    'mpx_helper.c',
>>    'seg_helper.c',
>> -  'smm_helper.c',
>>    'svm_helper.c',
>>    'tcg-cpu.c',
>>    'translate.c'), if_false: files('tcg-stub.c'))
>> diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
>> index 4ab30cc32e..35ba16dc3d 100644
>> --- a/target/i386/tcg/softmmu/meson.build
>> +++ b/target/i386/tcg/softmmu/meson.build
>> @@ -1,3 +1,4 @@
>>  i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
>>    'tcg-cpu.c',
>> +  'smm_helper.c',
>>  ))
>> 


-- 
Alex Bennée


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 12:32     ` Alex Bennée
@ 2021-02-15 12:59       ` Claudio Fontana
  2021-02-15 13:30         ` Paolo Bonzini
  0 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 12:59 UTC (permalink / raw)
  To: Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov, Paolo Bonzini,
	Philippe Mathieu-Daudé

On 2/15/21 1:32 PM, Alex Bennée wrote:
> 
> Claudio Fontana <cfontana@suse.de> writes:
> 
>> On 2/12/21 1:36 PM, Claudio Fontana wrote:
>>> smm is only really useful for softmmu, split in two modules
>>> around the CONFIG_USER_ONLY, in order to remove the ifdef
>>> and use the build system instead.
>>>
>>> Signed-off-by: Claudio Fontana <cfontana@suse.de>
>>> ---
>>>  target/i386/helper.h                       |  4 ++++
>>>  target/i386/tcg/seg_helper.c               |  2 ++
>>>  target/i386/tcg/{ => softmmu}/smm_helper.c | 19 ++-----------------
>>>  target/i386/tcg/translate.c                |  2 ++
>>>  target/i386/tcg/meson.build                |  1 -
>>>  target/i386/tcg/softmmu/meson.build        |  1 +
>>>  6 files changed, 11 insertions(+), 18 deletions(-)
>>>  rename target/i386/tcg/{ => softmmu}/smm_helper.c (98%)
>>>
>>> diff --git a/target/i386/helper.h b/target/i386/helper.h
>>> index c2ae2f7e61..8ffda4cdc6 100644
>>> --- a/target/i386/helper.h
>>> +++ b/target/i386/helper.h
>>> @@ -70,7 +70,11 @@ DEF_HELPER_1(clac, void, env)
>>>  DEF_HELPER_1(stac, void, env)
>>>  DEF_HELPER_3(boundw, void, env, tl, int)
>>>  DEF_HELPER_3(boundl, void, env, tl, int)
>>> +
>>> +#ifndef CONFIG_USER_ONLY
>>>  DEF_HELPER_1(rsm, void, env)
>>> +#endif /* !CONFIG_USER_ONLY */
>>> +
>>>  DEF_HELPER_2(into, void, env, int)
>>>  DEF_HELPER_2(cmpxchg8b_unlocked, void, env, tl)
>>>  DEF_HELPER_2(cmpxchg8b, void, env, tl)
>>> diff --git a/target/i386/tcg/seg_helper.c b/target/i386/tcg/seg_helper.c
>>> index 180d47f0e9..f0cb1bffe7 100644
>>> --- a/target/i386/tcg/seg_helper.c
>>> +++ b/target/i386/tcg/seg_helper.c
>>> @@ -1351,7 +1351,9 @@ bool x86_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
>>>      case CPU_INTERRUPT_SMI:
>>>          cpu_svm_check_intercept_param(env, SVM_EXIT_SMI, 0, 0);
>>>          cs->interrupt_request &= ~CPU_INTERRUPT_SMI;
>>> +#ifndef CONFIG_USER_ONLY
>>>          do_smm_enter(cpu);
>>> +#endif
>>>          break;
>>>      case CPU_INTERRUPT_NMI:
>>>          cpu_svm_check_intercept_param(env, SVM_EXIT_NMI, 0, 0);
>>> diff --git a/target/i386/tcg/smm_helper.c b/target/i386/tcg/softmmu/smm_helper.c
>>> similarity index 98%
>>> rename from target/i386/tcg/smm_helper.c
>>> rename to target/i386/tcg/softmmu/smm_helper.c
>>> index 62d027abd3..ee53b26629 100644
>>> --- a/target/i386/tcg/smm_helper.c
>>> +++ b/target/i386/tcg/softmmu/smm_helper.c
>>> @@ -1,5 +1,5 @@
>>>  /*
>>> - *  x86 SMM helpers
>>> + *  x86 SMM helpers (softmmu-only)
>>>   *
>>>   *  Copyright (c) 2003 Fabrice Bellard
>>>   *
>>> @@ -18,27 +18,14 @@
>>>   */
>>>  
>>>  #include "qemu/osdep.h"
>>> -#include "qemu/main-loop.h"
>>>  #include "cpu.h"
>>>  #include "exec/helper-proto.h"
>>>  #include "exec/log.h"
>>> -#include "helper-tcg.h"
>>> +#include "tcg/helper-tcg.h"
>>>  
>>>  
>>>  /* SMM support */
>>>  
>>> -#if defined(CONFIG_USER_ONLY)
>>> -
>>> -void do_smm_enter(X86CPU *cpu)
>>> -{
>>> -}
>>> -
>>> -void helper_rsm(CPUX86State *env)
>>> -{
>>> -}
>>> -
>>> -#else
>>> -
>>>  #ifdef TARGET_X86_64
>>>  #define SMM_REVISION_ID 0x00020064
>>>  #else
>>> @@ -330,5 +317,3 @@ void helper_rsm(CPUX86State *env)
>>>      qemu_log_mask(CPU_LOG_INT, "SMM: after RSM\n");
>>>      log_cpu_state_mask(CPU_LOG_INT, CPU(cpu), CPU_DUMP_CCOP);
>>>  }
>>> -
>>> -#endif /* !CONFIG_USER_ONLY */
>>> diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
>>> index af1faf9342..5075ac4830 100644
>>> --- a/target/i386/tcg/translate.c
>>> +++ b/target/i386/tcg/translate.c
>>> @@ -8321,7 +8321,9 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
>>>              goto illegal_op;
>>>          gen_update_cc_op(s);
>>>          gen_jmp_im(s, s->pc - s->cs_base);
>>> +#ifndef CONFIG_USER_ONLY
>>>          gen_helper_rsm(cpu_env);
>>> +#endif /* CONFIG_USER_ONLY */
>>>          gen_eob(s);
>>>          break;
>>
>> Hello Alex,
>>
>> this is something I wanted to bring in the foreground:
>>
>> while before we were generating an empty helper call for CONFIG_USER_ONLY,
>> now we are not generating anything.
> 
> Surely that says we only generate the helper call when we are not
> CONFIG_USER_ONLY?

Yes. The difference between before the patch and after the patch
is that before we were still going through all the code in tcg_gen_callN,
via the call to gen_helper_rsm macro, only to call finally an empty function for CONFIG_USER_ONLY (helper_rsm() {}),

while now we do not generate anything, we do not call the gen_helper_rsm macro at all, so we don't go through tcg_gen_callN.

> 
> 
>>
>>
>>
>>>      case 0x1b8: /* SSE4.2 popcnt */
>>> diff --git a/target/i386/tcg/meson.build b/target/i386/tcg/meson.build
>>> index 68fa0c3187..ec5daa1edc 100644
>>> --- a/target/i386/tcg/meson.build
>>> +++ b/target/i386/tcg/meson.build
>>> @@ -8,7 +8,6 @@ i386_ss.add(when: 'CONFIG_TCG', if_true: files(
>>>    'misc_helper.c',
>>>    'mpx_helper.c',
>>>    'seg_helper.c',
>>> -  'smm_helper.c',
>>>    'svm_helper.c',
>>>    'tcg-cpu.c',
>>>    'translate.c'), if_false: files('tcg-stub.c'))
>>> diff --git a/target/i386/tcg/softmmu/meson.build b/target/i386/tcg/softmmu/meson.build
>>> index 4ab30cc32e..35ba16dc3d 100644
>>> --- a/target/i386/tcg/softmmu/meson.build
>>> +++ b/target/i386/tcg/softmmu/meson.build
>>> @@ -1,3 +1,4 @@
>>>  i386_softmmu_ss.add(when: ['CONFIG_TCG', 'CONFIG_SOFTMMU'], if_true: files(
>>>    'tcg-cpu.c',
>>> +  'smm_helper.c',
>>>  ))
>>>
> 
> 



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 12:59       ` Claudio Fontana
@ 2021-02-15 13:30         ` Paolo Bonzini
  2021-02-15 14:05           ` Claudio Fontana
  0 siblings, 1 reply; 33+ messages in thread
From: Paolo Bonzini @ 2021-02-15 13:30 UTC (permalink / raw)
  To: Claudio Fontana, Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov,
	Philippe Mathieu-Daudé

On 15/02/21 13:59, Claudio Fontana wrote:
> Yes. The difference between before the patch and after the patch
> is that before we were still going through all the code in tcg_gen_callN,
> via the call to gen_helper_rsm macro, only to call finally an empty function for CONFIG_USER_ONLY (helper_rsm() {}),
> 
> while now we do not generate anything, we do not call the gen_helper_rsm macro at all, so we don't go through tcg_gen_callN.
> 

Can we even have an abort() for such cases?

Paolo



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 13:30         ` Paolo Bonzini
@ 2021-02-15 14:05           ` Claudio Fontana
  2021-02-15 14:13             ` Paolo Bonzini
  0 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 14:05 UTC (permalink / raw)
  To: Paolo Bonzini, Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov,
	Philippe Mathieu-Daudé

On 2/15/21 2:30 PM, Paolo Bonzini wrote:
> On 15/02/21 13:59, Claudio Fontana wrote:
>> Yes. The difference between before the patch and after the patch
>> is that before we were still going through all the code in tcg_gen_callN,
>> via the call to gen_helper_rsm macro, only to call finally an empty function for CONFIG_USER_ONLY (helper_rsm() {}),
>>
>> while now we do not generate anything, we do not call the gen_helper_rsm macro at all, so we don't go through tcg_gen_callN.
>>
> 
> Can we even have an abort() for such cases?
> 
> Paolo
> 

Hi Paolo,

where are you suggesting to have an abort()?

You mean that we should abort() QEMU as soon as we detect in translate.c an RSM instruction in user-mode?

Again the translate.c code for reference:

    case 0x1aa: /* rsm */
        gen_svm_check_intercept(s, pc_start, SVM_EXIT_RSM);
        if (!(s->flags & HF_SMM_MASK))
            goto illegal_op;
        gen_update_cc_op(s);
        gen_jmp_im(s, s->pc - s->cs_base);
#ifndef CONFIG_USER_ONLY
        gen_helper_rsm(cpu_env);
#endif /* CONFIG_USER_ONLY */
        gen_eob(s);
        break;

---

Thanks,

CLaudio

  


^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 14:05           ` Claudio Fontana
@ 2021-02-15 14:13             ` Paolo Bonzini
  2021-02-15 14:39               ` Claudio Fontana
  0 siblings, 1 reply; 33+ messages in thread
From: Paolo Bonzini @ 2021-02-15 14:13 UTC (permalink / raw)
  To: Claudio Fontana, Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov,
	Philippe Mathieu-Daudé

On 15/02/21 15:05, Claudio Fontana wrote:
> On 2/15/21 2:30 PM, Paolo Bonzini wrote:
>> On 15/02/21 13:59, Claudio Fontana wrote:
>>> Yes. The difference between before the patch and after the patch 
>>> is that before we were still going through all the code in
>>> tcg_gen_callN, via the call to gen_helper_rsm macro, only to call
>>> finally an empty function for CONFIG_USER_ONLY (helper_rsm()
>>> {}),
>>> 
>>> while now we do not generate anything, we do not call the
>>> gen_helper_rsm macro at all, so we don't go through
>>> tcg_gen_callN.
>>> 
>> 
>> Can we even have an abort() for such cases?
>> 
>> Paolo
>> 
> 
> Hi Paolo,
> 
> where are you suggesting to have an abort()?
> 
> You mean that we should abort() QEMU as soon as we detect in
> translate.c an RSM instruction in user-mode?

Translating it is okay (it's just a guaranteed SIGILL), but I'm thinking
of aborting if s->flags & HF_SMM_MASK is true.  Likewise if we see
CPU_INTERRUPT_SMI.

Paolo

> 
>     case 0x1aa: /* rsm */
>         gen_svm_check_intercept(s, pc_start, SVM_EXIT_RSM);
>         if (!(s->flags & HF_SMM_MASK))
>             goto illegal_op;
>         gen_update_cc_op(s);
>         gen_jmp_im(s, s->pc - s->cs_base);
> #ifndef CONFIG_USER_ONLY
>         gen_helper_rsm(cpu_env);
> #endif /* CONFIG_USER_ONLY */
>         gen_eob(s);
>         break;



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 14:13             ` Paolo Bonzini
@ 2021-02-15 14:39               ` Claudio Fontana
  2021-02-15 15:33                 ` Claudio Fontana
  0 siblings, 1 reply; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 14:39 UTC (permalink / raw)
  To: Paolo Bonzini, Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov,
	Philippe Mathieu-Daudé

On 2/15/21 3:13 PM, Paolo Bonzini wrote:
> On 15/02/21 15:05, Claudio Fontana wrote:
>> On 2/15/21 2:30 PM, Paolo Bonzini wrote:
>>> On 15/02/21 13:59, Claudio Fontana wrote:
>>>> Yes. The difference between before the patch and after the patch 
>>>> is that before we were still going through all the code in
>>>> tcg_gen_callN, via the call to gen_helper_rsm macro, only to call
>>>> finally an empty function for CONFIG_USER_ONLY (helper_rsm()
>>>> {}),
>>>>
>>>> while now we do not generate anything, we do not call the
>>>> gen_helper_rsm macro at all, so we don't go through
>>>> tcg_gen_callN.
>>>>
>>>
>>> Can we even have an abort() for such cases?
>>>
>>> Paolo
>>>
>>
>> Hi Paolo,
>>
>> where are you suggesting to have an abort()?
>>
>> You mean that we should abort() QEMU as soon as we detect in
>> translate.c an RSM instruction in user-mode?
> 
> Translating it is okay (it's just a guaranteed SIGILL), but I'm thinking
> of aborting if s->flags & HF_SMM_MASK is true.  Likewise if we see
> CPU_INTERRUPT_SMI.
> 
> Paolo
> 

Ok, will rework as you suggest, thanks!

>>
>>     case 0x1aa: /* rsm */
>>         gen_svm_check_intercept(s, pc_start, SVM_EXIT_RSM);
>>         if (!(s->flags & HF_SMM_MASK))
>>             goto illegal_op;
>>         gen_update_cc_op(s);
>>         gen_jmp_im(s, s->pc - s->cs_base);
>> #ifndef CONFIG_USER_ONLY
>>         gen_helper_rsm(cpu_env);
>> #endif /* CONFIG_USER_ONLY */
>>         gen_eob(s);
>>         break;
> 
> 



^ permalink raw reply	[flat|nested] 33+ messages in thread

* Re: [RFC v18 08/15] i386: split smm helper (softmmu)
  2021-02-15 14:39               ` Claudio Fontana
@ 2021-02-15 15:33                 ` Claudio Fontana
  0 siblings, 0 replies; 33+ messages in thread
From: Claudio Fontana @ 2021-02-15 15:33 UTC (permalink / raw)
  To: Paolo Bonzini, Alex Bennée
  Cc: Laurent Vivier, Peter Maydell, Thomas Huth, Eduardo Habkost,
	Richard Henderson, qemu-devel, Roman Bolshakov,
	Philippe Mathieu-Daudé

On 2/15/21 3:39 PM, Claudio Fontana wrote:
> On 2/15/21 3:13 PM, Paolo Bonzini wrote:
>> On 15/02/21 15:05, Claudio Fontana wrote:
>>> On 2/15/21 2:30 PM, Paolo Bonzini wrote:
>>>> On 15/02/21 13:59, Claudio Fontana wrote:
>>>>> Yes. The difference between before the patch and after the patch 
>>>>> is that before we were still going through all the code in
>>>>> tcg_gen_callN, via the call to gen_helper_rsm macro, only to call
>>>>> finally an empty function for CONFIG_USER_ONLY (helper_rsm()
>>>>> {}),
>>>>>
>>>>> while now we do not generate anything, we do not call the
>>>>> gen_helper_rsm macro at all, so we don't go through
>>>>> tcg_gen_callN.
>>>>>
>>>>
>>>> Can we even have an abort() for such cases?
>>>>
>>>> Paolo
>>>>
>>>
>>> Hi Paolo,
>>>
>>> where are you suggesting to have an abort()?
>>>
>>> You mean that we should abort() QEMU as soon as we detect in
>>> translate.c an RSM instruction in user-mode?
>>
>> Translating it is okay (it's just a guaranteed SIGILL), but I'm thinking
>> of aborting if s->flags & HF_SMM_MASK is true.  Likewise if we see
>> CPU_INTERRUPT_SMI.
>>
>> Paolo
>>
> 
> Ok, will rework as you suggest, thanks!


By the way, in the case of gen_bpt_io, is it a similar situation,
where we should abort in user-mode if we see s->flags & HF_IOBPT_MASK ?


static void gen_bpt_io(DisasContext *s, TCGv_i32 t_port, int ot)
{
#ifndef CONFIG_USER_ONLY
    if (s->flags & HF_IOBPT_MASK) {
        TCGv_i32 t_size = tcg_const_i32(1 << ot);
        TCGv t_next = tcg_const_tl(s->pc - s->cs_base);

        gen_helper_bpt_io(cpu_env, t_port, t_size, t_next);
        tcg_temp_free_i32(t_size);
        tcg_temp_free(t_next);
    }
#endif /* !CONFIG_USER_ONLY */
}



What about other cases like


        case 0xd8: /* VMRUN */

            if (!(s->flags & HF_SVME_MASK) || !s->pe) {
                goto illegal_op;
            }

          ...

            gen_helper_vmrun(cpu_env, tcg_const_i32(s->aflag - 1),
                             tcg_const_i32(s->pc - pc_start));


should we abort there as well if CONFIG_USER_ONLY?

And there are many more probably, should it be its own patch?

Ciao,

Claudio

> 
>>>
>>>     case 0x1aa: /* rsm */
>>>         gen_svm_check_intercept(s, pc_start, SVM_EXIT_RSM);
>>>         if (!(s->flags & HF_SMM_MASK))
>>>             goto illegal_op;
>>>         gen_update_cc_op(s);
>>>         gen_jmp_im(s, s->pc - s->cs_base);
>>> #ifndef CONFIG_USER_ONLY
>>>         gen_helper_rsm(cpu_env);
>>> #endif /* CONFIG_USER_ONLY */
>>>         gen_eob(s);
>>>         break;
>>
>>
> 
> 



^ permalink raw reply	[flat|nested] 33+ messages in thread

end of thread, other threads:[~2021-02-15 15:34 UTC | newest]

Thread overview: 33+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-12 12:36 [RFC v18 00/15] i386 cleanup PART 2 Claudio Fontana
2021-02-12 12:36 ` [RFC v18 01/15] i386: split cpu accelerators from cpu.c, using AccelCPUClass Claudio Fontana
2021-02-15 11:29   ` Alex Bennée
2021-02-12 12:36 ` [RFC v18 02/15] cpu: call AccelCPUClass::cpu_realizefn in cpu_exec_realizefn Claudio Fontana
2021-02-15 11:30   ` Alex Bennée
2021-02-12 12:36 ` [RFC v18 03/15] accel: introduce new accessor functions Claudio Fontana
2021-02-15 11:34   ` Alex Bennée
2021-02-12 12:36 ` [RFC v18 04/15] target/i386: fix host_cpu_adjust_phys_bits error handling Claudio Fontana
2021-02-12 12:36 ` [RFC v18 05/15] accel-cpu: make cpu_realizefn return a bool Claudio Fontana
2021-02-12 12:36 ` [RFC v18 06/15] meson: add target_user_arch Claudio Fontana
2021-02-15 11:37   ` Alex Bennée
2021-02-12 12:36 ` [RFC v18 07/15] i386: split off softmmu-only functionality in tcg-cpu Claudio Fontana
2021-02-12 12:36 ` [RFC v18 08/15] i386: split smm helper (softmmu) Claudio Fontana
2021-02-15 11:51   ` Claudio Fontana
2021-02-15 12:32     ` Alex Bennée
2021-02-15 12:59       ` Claudio Fontana
2021-02-15 13:30         ` Paolo Bonzini
2021-02-15 14:05           ` Claudio Fontana
2021-02-15 14:13             ` Paolo Bonzini
2021-02-15 14:39               ` Claudio Fontana
2021-02-15 15:33                 ` Claudio Fontana
2021-02-12 12:36 ` [RFC v18 09/15] i386: split tcg excp_helper into softmmu and user parts Claudio Fontana
2021-02-12 12:36 ` [RFC v18 10/15] i386: split tcg btp_helper " Claudio Fontana
2021-02-15 11:55   ` Claudio Fontana
2021-02-12 12:36 ` [RFC v18 11/15] i386: split misc helper into user and softmmu parts Claudio Fontana
2021-02-12 12:36 ` [RFC v18 12/15] i386: separate fpu_helper " Claudio Fontana
2021-02-15 10:32   ` Alex Bennée
2021-02-12 12:36 ` [RFC v18 13/15] i386: slit svm_helper into softmmu and stub-only user Claudio Fontana
2021-02-12 12:36 ` [RFC v18 14/15] i386: split seg_helper into user-only and softmmu parts Claudio Fontana
2021-02-12 12:36 ` [RFC v18 15/15] i386: split off softmmu part of cpu.c Claudio Fontana
2021-02-12 12:57 ` [RFC v18 00/15] i386 cleanup PART 2 no-reply
2021-02-15 11:37 ` Alex Bennée
2021-02-15 11:48   ` Claudio Fontana

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).